LWD

LWD - 07 FIRST INTERNATIONAL CONFERENCE ON TECHNOLOGY-BASED LEARNING WITH DISABILITY

July 19-20, 2007 Wright State University Dayton, Ohio

Proceedings

LWD – 07 FIRST INTERNATIONAL CONFERENCE ON TECHNOLOGYBASED LEARNING WITH DISABILITY

JULY 19-20, 2007 WRIGHT STATE UNIVERSITY DAYTON, OHIO

SPONSORED BY: THE NATIONAL SCIENCE FOUNDATION AND WRIGHT STATE UNIVERSITY

IN COOPERATION WITH: GOODWILL EASTER SEALS AND UNITED REHABILITATION SERVICES

ii

TABLE OF CONTENTS First International Conference on Technology-based Learning with Disability Chairs’ Message………………………………..…………........………......………………….....vii Organizing and Technical Program Committees………..……………..……………..................viii Special Presentations…………………………………………………....…………..……………ix Acknowledgements………………………………………………………………………………..x Session 1: STEM Learning with Disability Reducing Multiple Interpretations of Mathematical Expressions with MathSpeak Mickey Isaacson, Lyle Lloyd and Dave Schleppenbach Purdue University & gh LLC…………………….........................................................................................................1 Universal Design for Math Learning: Bridging the Technology and Policy Divide Steve Noble, Design Science, Inc., Louisville, KY…………………...............................................7 Novel Approaches to Deaf Education Nicoletta Adamo-Villani and Ronnie Wilbur, Purdue University………………...……………..13 AutoMathic Blocks-An Automated System to Teach Math to K-12 Children with Severe Visual Impairment Allowing Both Physical and Haptic Interaction with an Automated Tutor Arthur Karshmer, Priyanka Daultani and Michael McCaffrey, University of San Francisco; Abdel Ejnioui,University of South Florida and Roope Raisamo, University of Tampere Finland………………………………….…………………………………………...22 AccessScope Project: Accessible Light Microscopy for Students with Disabilities Bradley Duerstock, Wamiq Ahmed, John Cirillo and J. Paul Robinson, Purdue University………………………………………………………………………………………...30 Using Cosmo's Learning System (CLS) with Children with Autism Corinna Lathan, Katharina Boser, Charlotte Safos, Cathy Frentz and Kaitlin Powers, AnthroTronix, Inc.,Silver Springs, MD…………………………………………….……….……37

iii

Session 2: Human-Centered Technology Concepts in LWD – I Studies on Display/Icon Complexity and Its Relationship to Performance D.Repperger, D. Aleva, G. Thomas, Air Force Research Laboratory; S. Fullenkamp, General Dynamics and C. Phillips, Wright State University……………………………………………...48 Gaze Pointing and facial EMG Clicking Andrew Junker, Brain Actuated Technologies, Inc., Yellow Springs, OH and John Paulin Hansen, IT University, Copenhagen…………………………………..…………………………56 Computer Access Using Electrical Signals from the Forehead: The Cyberlink in Action Julio Mateo and Markus Feufel, Wright State University……………………………………….62 IBIS: Intelligent Binary Information Selection Katherine McCreight, N-Space Analysis, LLC, Xenia, OH…………………...…………………68 Enhancing Learning with Haptic Devices as a Possible Assistive Aid D. Repperger, Air Force Research Laboratory and C. Phillips, Wright State University………………………………………………………………………..……………….76 Session 3: Accessible Curricula and Teaching for LWD Delivering on the Promise of Plato's Academy: Accessible STEM curricula for the Universitas Scholarium of the 21st Century Michele Wheatly, Forouzan Golshani and Jeffrey Vernooy, Wright State University………………………………………………………………………………...............84 Connecting Teaching and Learning Greg Stefanich, University of Northern Iowa……………………………………………………92 Inclusive Science Education: Classroom Teacher & Science Educator Experiences in CLASS Susan Kirch, City University of New York; Mary Ellen Bargerhuff and Heidi Cowan, Wright State University…………………………………………………………………………………102 Descriptions of the Science, Technology, Engineering and Math (STEM) Workshops for Middle School Students with Specific Learning Disabilities Jiang Zhe, Julie Zhao, Paul Lam, Dennis Doverspike and Craig Menzemer, University of Akron, OH………………………………………………………………................................................112 Equity in Accessing Academic Content: Support for General Educators Teaching Students with Disabilities in Inclusive Classrooms Heidi Cowan, Mary Ellen Bargerhuff and Jackie Collier, Wright State University..………….121

iv

Session 4: Information Access and On-Line LWD Accessibility of Higher Education Websites in the Northwestern U.S.: Current Status and Response to Third Party Outreach Terry Thompson, Sheryl Burgstahler and Elizabeth Moore, University of Washington…………………………………………………………………………..…………127 Software Helping Military Troops and Persons with Disabilities Help One Another Wayne Shebilske, Shruti Narakesari, Ganesh Alakke and John Kegley, Wright State University………………………………………………………………………….....................137 OASIS System for Organizing, Annotating, and Serving Information to Students without Sight Qing Li, Selcuk Candan, Maria Goveas, Sangwoo Han,Terri Hedgpeth, Jong Wook Kim, Atul Kolhatkar and Maria Luisa Sapino, Arizona State University…………………........................144 The Universal eLearner-An Innovative Approach for Universal Online Learning for all Students Barry Cronin and Matt Kaplowitz, Bridge Multimedia, New York, NY; Caesar Eghtesadi, American Foundation for the Blind, New York, NY………………….........................................153

Bridging the Digital Divide: Linking Printed Pages to Digital Content for Distance Learning Applications. Michael Mott, Saint Xavier University and Jason Barkeloo, Somatic Digital, Cincinnati, OH …………………………………………………………………………………………..………168 Bridging the Digital Divide to Benefit Mentally III Clients Carol Buechler, Buechler Computers and Engineers, Dayton, OH……………...…………….181 Session 5: Human-Centered Technology Concepts in LWD – II Multi Sensory Books - Assistive Technology Meets 21st Century Book Publishing Barry Cronin and Matt Kaplowitz, Bridge Multimedia, New York, NY; Judy Vesel, TERC, Inc., Cambridge, MA…………………………………………………………………………………187 Measuring Psychological Adjustment to Disability Allan Dodds, Nottingham UK…………………………………………………………………..197 Sense Making in Disability Management Katherine Lippa and Valerie Shalin, Wright State University…………………...…………….206 Development of a universal interactive display for three-dimensional cartographic information: The Smithsonian Institution's Talking Touch Model of the Washington, D.C. National Mall and Environs Steven Landau, Touch Graphics, Inc., New York, NY………………………………………….214

v

Virtual Reality Environment to Assist Disabled Individuals Maurissa D'Angelo, Wright State University…………………………………………………..221 Session 6: Learning with Communication Disabilities Equalizing Sign and Spoken Language in the Chemistry Laboratory, Brenda Seal, Gina MacDonald and Daniel Downey, James Madison University, Harrisonburg, VA ……………………………………………………………………………………………....229 Opportunities and Challenges of Using Technology in Deaf Education Kent Robertson, The Shodor Education Foundation, Inc.and Jana Lollis, North Carolina School for the Deaf; Morganton, NC…………………………………………………………………..234 Voice-to-Text Realtime Transcription: A Cost-Effective Interpretation System for Deaf and Hard of Hearing Students Michael Kress, Margaret Venditti, Maryellen Smolka, Nicole Dory, Sheryll Porter and Joseph Nicolosi, City University of New York………………………………………………………….242 Access to Library Collections by Gaze Interaction Haakon Lund, Institute of Information Studies, Royal School of Library and Information Science, Copenhagen Denmark and John Paulin Hansen, IT University of Copenhagen, Denmark………………………………………………………………………………...............253 Walking the Path with the Twice Exceptional (2E) Learner: Understanding the Paradox of Exceptional Strengths and Weaknesses Katharina Boser and Trish Budd, Individual Differences in Learning Association……………260 Alternative and Augmented Communication Device Feature Matching Brandi-Lynn Greig, United Rehabilitation Services, Dayton, OH……………………………..267

vi

Chairs’ Message Message from the LWD-07 Chairs The concept of conducting a conference on Technology-based Learning with Disability was conceived as an important means of bringing together a broad spectrum of researchers and practitioners with common interests in human disabilities to exchange their ideas and findings. As the full title implies, LWD-07 is broadly focused on interdisciplinary approaches to facilitating formal education and life-long learning for people with all types of disabilities. Examination of the contents of this proceedings document offers ample evidence of our success in achieving the goals of geographical and intellectual diversity within the confines a relatively short inaugural conference. Papers from across the United States and Europe, and topics ranging from advanced display and control interfaces to national policy on disability and accommodations for science and technology education, are represented in the mix. The key supporters of LWD-07 were Wright State University and the National Science Foundation. Celebrating its 40th anniversary as an independent university in 2007, Wright State continues as an innovator in architectural and curricular accessibility for students with disabilities. NSF is a major funding agency for a variety of programs that foster educational endeavors for persons with disabilities including the Integrative Graduate Education and Research Traineeship (IGERT) Ph.D. concentration in Technology-based Learning with Disability at Wright State University. Forouzan Golshani, Michele Wheatly – General Chairs Clark Shingledecker – Program Chair

vii

Organizing Committee Forouzan Golshani and Michele Wheatly – General Chairs Clark Shingledecker – Program Chair Katherine Myers – Exhibits Chair

Technical Program Committee Mary Ellen Bargerhuff, Wright State University Nikolaos Bourbakis, Wright State University Sheryl Burgstahler, University of Washington Allan Dodds, Consultant Health/Neuropsychologist, UK John Flach, Wright State University Robert Fyffe, Wright State University Forouzan Golshani, Wright State University John Paulin Hansen, IT University Copenhagen, Denmark Andrew Junker, Brain Actuated Technologies Inc. David MacKay, Cambridge University, UK S. Panch Panchanathan, Arizona State University Daniel Repperger, Air Force Research Laboratory David Reynolds, Wright State University Barry Romich, Prenke Romich Company Clark Shingledecker, Wright State University Virginia Stern, American Association for the Advancement of Science Jeff Vernooy, Wright State University Michele Wheatly, Wright State University

viii

Special Presentations

Keynote: Day 1 Leaving the State of Stuck: Technologies for Being Connected and Getting Around. Dr. Marcia Scherer, Institute for Matching Person and Technology

Keynote: Day 2 Advancing Research on STEM Learning with Disability. Dr. Sharon Locke, Program Director, NSF Division of Research on Learning

Panel Presentation – Day 2 40 Years of Accessible Education. Bill Rickert, WSU Associate Provost - Panel Chair Jeffery Vernooy, Stephen Simon and Pat Marx – Current and past directors, Office of Disability Services, Wright State University

ix

Acknowledgements Special thanks are extended to the following individuals for their invaluable contributions to the success of LWD-07:

Cynthia Boone, Wright State University -- Administrative and program support John Flach, Wright State University -- Technical session chair Stephen Fortson, Wright State University – Technical Session Chair Anne Johnson, Mayer Johnson, Inc. – Technical session chair Jeffrey Vernooy, Wright State University – Conference planning, disability services coordination, technical session chair Mark Willis, Wright State University – Technical Session Chair Youngchoon Park, Johnson Controls Incorporated -- Conference proceedings production, technical session chair

x

First International Conference on Technology-based Learning with Disability

Reducing Multiple Interpretations of Mathematical Expressions with MathSpeak Mickey D. Isaacson, Lyle L. Lloyd, & Dave Schleppenbach Purdue University & gh LLC West Lafayette, Indiana ABSTRACT This was a preliminary efficacy study to examine the capacity of MathSpeak for reducing multiple interpretations inherent in spoken mathematical expressions. Multiple interpretations inhibit the learning of mathematics by print-disabled individuals (e.g., blindness, low-vision, learning disabilities) who rely on spoken communication. Twenty-two million Americans have disabilities that typically prevent reading ordinary print. These individuals are not competitive in today's high-tech, information-laden society. In particular, fields of science, technology, engineering, and mathematics (STEM) are often closed to these individuals. One barrier in providing print-disabled people with equal access to STEM fields involves difficulty accurately conveying mathematical knowledge because of ambiguities in spoken communication of mathematical expressions. Speech, which is frequently used to transmit mathematical information, is problematic because of ambiguities inherent in spoken language. Print-disabled individuals rely heavily on speech for receiving information. Although there have been several conference presentations on how to disambiguate spoken mathematical expression, the authors know of no peer-reviewed publications (or conference presentations) demonstrating the efficacy of these approaches for reducing multiple interpretations. MathSpeak is a product under development containing a standardized set of rules (based on the Nemeth code) for presenting math expressions in a non-ambiguous format. The purpose of the present study was to examine MathSpeak’s capacity for disambiguation. Twenty-eight students participated in a within-subjects efficacy study. Results show MathSpeak is highly efficacious. Significantly more expressions heard with the current version of MathSpeak were correctly interpreted compared to expressions heard in a form typical of everyday spoken language. INTRODUCTION Twenty-two million Americans have disabilities (blindness/low-vision, learning/cognitive disabilities) that prevent reading ordinary print (Census data on people with disabilities, 2002; U.S. Census Bureau, 2003). These individuals (a.k.a., print-disabled) are not competitive in today's high-tech, information-laden society. Fields of science, technology, engineering, and mathematics (STEM) are often closed to them. Due to this lack of access, the unemployment rate among print-disabled individuals is more than three times their non-disabled counterparts nationwide (U.S. Department of Health and Human Services, n.d.). Various reports (The Center for an Accessible Society, n.d.; In

1

First International Conference on Technology-based Learning with Disability Unison2000, 2000) put the unemployment rate among disabled individuals between 30%-60%. People with disabilities are, also, significantly under represented in STEM related fields (Burstahler, 1994; Malcom, S. M. & Matyas, M. L., 1991). Similar under-representation is apparent in colleges and universities (National Science Foundation Committee on Equal Opportunities in Science and Engineering, 2000). Most unemployed disabled are receiving aid under both state and federal programs. In 2002, Indiana spent over $70 million to assist 33,257 people with disabilities to find employment, but only 3,980 were able to secure at least part-time employment (Indiana Family and Social Services Administration, n.d.). Unemployment/ underemployment is a huge government/taxpayer burden. Individuals with disabilities comprise 10.4% of the overall workforce, but only 2.7% of the science and engineering workforce. This gap is not necessarily indicative of a lack of interest in STEM careers. According to the American Council on Education (ACE), college freshmen with disabilities are equally interested in majoring in science as their non-disabled counterparts, however, this initial interest is rarely realized as an actual career in a STEM-related field (Henderson, 1999). Further evidence of the shortage of individuals with disabilities in STEM fields comes from a National Science Foundation study finding less than 320 persons with disabilities received Ph.D.s in Science/Engineering in 1997 (National Science Foundation, 2000). One problem in providing print-disabled people with equal access to STEM fields involves accurately conveying mathematical knowledge. Speech, which is frequently used, is problematic because of inherent ambiguities. To illustrate, consider the following algebraic expression: 1 + 2 . Typically, this expression is read aloud as: “one over one plus two plus 1+ 2

two” but it is ambiguous with multiple interpretations (specifically, and

1 ). 1+ 2 + 2

1 1 +2, + 2+ 2, 1+ 2 1

This example is only one of a multitude of similar ambiguities.

Another source inhibiting equal access involves rapidly producing audio renderings of mathematically-based material. Current conventional audio interventions for print-disabled students rely on the assistance of human readers or laborious by-hand human recording onto cassette tapes of tests, training manuals and other informational sources. Due to the expense and time-consuming nature of this work, many students receive books 4-6 months after the print versions are available and others never receive any books. Moreover, productive employment by print-disabled individuals is hindered by the lack of rapid rendering of job-related mathematical material. A preliminary version of a product called MathSpeak for the disambiguation of oral presentations of mathematical expressions is under development at gh LLC in Purdue’s Research Park. The previous example of 1 + 2 would be expressed orally in MathSpeak as “begin 1+ 2

fraction, one over one plus two, end fraction, plus two.” This rendering has only one possible interpretation. Additionally, the MathSpeak product contains a computerized component that easily and rapidly translates STEM materials into the non-ambiguous MathSpeak form and then produces a computer-synthesized auditory rendering. MathSpeak has great potential for reducing ambiguities in auditory presentations of STEM materials in a time-efficient manner, however, no research has been conducted to determine the efficacy of this product. Hence, the purpose of the present study was to examine the capacity of MathSpeak for disambiguation.

2

First International Conference on Technology-based Learning with Disability METHOD Twenty-eight undergraduate education majors (10 males; 18 females) served as participants. In the experimental condition, participants received training explaining how the MathSpeak terminology would be used to present audio rendering of mathematical expressions. Next, they listened to mathematical expressions presented auditorily with the Mathspeak terminology and were asked to circle from four possible choices, the expression intended to be conveyed through the audio presentation. In the control condition, participants received training explaining how mathematical expressions would be presented through audio renderings typical of common everyday speech. After training, they listened to mathematical expressions presented in common speech renderings and were asked to circle the choice they believed was intended to be conveyed. Each participant was tested in control and MathSpeak conditions. Half of the participants received the MathSpeak condition first followed by the control; the other half received the reverse order. RESULTS A 2 x 2 x 2 mixed factorial analysis of variance was performed on the number of expressions correctly interpreted. Between subject factors were gender and order of tests. Terminology was within subject. Main effects of terminology, F(1,48) = 197.689, p < .001, and gender, F(1,48) = 6.970, p = .011 were found. These data are summarized in Figure 1.

Figure 1 – Number of expressions interpreted correctly for the two types of terminology The main effect of terminology is evident in the greater number of expressions correctly interpreted when MathSpeak terminology was used compared to common terminology (mean scores collapsed over sex were 18.93 for MathSpeak and 8.07 for common). In fact, the distributions of scores for the two types of terminology were non-overlapping. A t-test modified for differences between proportions found correctly interpreting expressions with common 3

First International Conference on Technology-based Learning with Disability terminology did not differ significantly from chance. The main effect of gender is evident in a slightly higher number of expressions correctly interpreted for males relative to females (mean scores collapsed over terminology types for males and females were 15.00 and 13.17 respectively). Prior mathematical knowledge as determined by the number of courses taken during high school and college is displayed in Figure 2. The majority of participants had completed two or more math courses. Mean, median, and mode for the number of prior courses were 4.64 (SD +1.37); 5; and 5, respectively. No significant correlations were found between the number of math courses taken and the number of expressions correctly interpreted for either type of terminology. Knowledge Base

# of Participants

10 8 6 4 2 0 1

2

3

4

5

6

7

8

# of Math Courses

Figure 2- Distribution of prior math courses DISCUSSION Federal legislation (IDEA, 2004) mandates all students shall have a free and appropriate education. For students with print disabilities, lack of accessibility to printed materials creates a barrier for providing an appropriate education. To overcome this barrier, IDEA (2004) emphasizes the need for a National Instructional Materials Accessibility Standard (NIMAS) for providing instructional materials in a timely and unambiguous manner to blind and other persons with print disabilities. Based on the present study, MathSpeak appears to be efficacious in reducing multiple interpretations. It was found that correctly interpreting ambiguous mathematical expression with commonly used terminology did not differ significantly from chance performance, however, when MathSpeak terminology was implemented, performance increased significantly to ceiling levels of approximately 95% of expressions correctly interpreted. MathSpeak’s capacity for presenting audio expressions in an unambiguous manner should facilitate the achievement of the NIMAS objectives and entry in to STEM fields by individuals with disabilities. A potential factor that could influence perceptual processes related to auditory input of mathematical information is prior amount of math experience. The number of math courses taken prior to testing was examined for possible correlations with performance for both

4

First International Conference on Technology-based Learning with Disability terminology types and no relationships were found. Although these data suggest that this factor does not influence performance, it is important to temper any conclusion with the fact that all undergraduates entering college need to have a prerequisite number of math courses, resulting in a small range of courses and a concomitant decrease in sensitivity for correlation testing. The effect of gender was an unexpected finding. Most likely the effect is an artifact that has arisen from: 1) the smaller sample size of males relative to females; and 2) the relatively small number of males overall. Participant sampling was based on solicitation from an undergraduate education course and the 2 to 1 ratio of females to males is representative of most education classes. Inclusion was based on availability and no attempt was made to equate numbers based on gender. Future studies will need to employ sampling techniques that equate the number of males and females to be tested. Despite, this unexpected finding, the nonoverlapping distributions for the two terminology types clearly establishes the efficacy of MathSpeak as a vehicle for enhancing disambiguation of audio renderings of mathematical expressions. REFERENCES Burstahler, S. (1994). Increasing the representation of people with disabilities in science, engineering, and mathematics. Information Technology and Disability 1(4). Retrieved November 16, 2006 from http://www.washington.edu/doit/Brochures/Careers/representation.html Census data on people with disabilities interesting facts. (2002). Retrieved November 16, 2006 from http://www.ilru.org/healthwellness/html/census.html Indiana Family and Social Services Administration. Division of Disability and Rehabilitative Services Reports and Statistics. Retrieved November 16, 2006 from http://www.in.gov/fssa/disability/statistics/index.html Individuals with Disabilities Education Act (IDEA) of 2004, Pub. L. No. 108-446, § 118, Stat. 2647 (2004). Henderson, C. (1999) College freshmen with disabilities: A biennial Statistical Profile. American Council on Education, Washington, DC, 1999. In Unison2000. (2000). Retrieved November 16, 2006 from http://socialunion.gc.ca/In_Unison2000/iu03100e.html Malcom, S. M., & Matyas, M. L. (1991). Investing in human potential: Science and engineering at the crossroads. Washington, D. C.: American Association for the Advancement of Science. National Science Foundation (2000). Women, minorities, and persons with disabilities in science and engineering: 2000 (NSF-0327); Arlington: VA. National Science Foundation Committee on Equal Opportunities in Science and Engineering. (2000). Biennial Report to the United States Congress. Retrieved November 16, 2006 from http://www.nsf.gov/pubs/2001/ceose2000rpt/congress_5.pdf The Center for an Accessible Society. Retrieved April 26, 2007 from http://www.accessiblesociety.org/topics/economics-employment/labor2002.html U.S Census Bureau. (2003). U.S. census data. Retrieved November 16, 2006 from http://www.census.gov U.S. Department of Health and Human Services. Retrieved November 16, 2006 from http://www.cms.hhs.gov/center/people.asp

5


Acknowledgements Special thanks to: 1) Eric Graves, Kelly Hough, Samuel Mathew, & Jonathan Williford for assistance with methodological development, data collection, and final preparation of the manuscript; 2) gh LLC for supplying programming expertise, materials, and financial resources for participant compensation; 3) the Purdue AAC seminar group for constructive feedback; and 4) Purdue’s statistical consulting services for data analysis advice.

6


Universal Design for Math Learning: Bridging the Technology and Policy Divide Steve Noble Director of Accessibility Policy Design Science, Inc. (502) 969-3088 [email protected] ABSTRACT This paper attempts to make fundamental connections between the technological capabilities now available for creating universally designed math, and the body of public policy which demands that educational offerings be made accessible to students with disabilities. The premise to be examined is that making math accessible is as much a public policy issue as it is a technological one. The concept of what is required under law (public policy) continues to expand as the technological issues of effective access are resolved. Now that the technological issues of accessible math have been resolved, it is essential that disability advocates and educators push for better public policy to support the availability of accessible math in the classroom. CONNECTING MATH PERFORMANCE AND ACCESSIBILITY The attainment of good math skills has been identified as one of the major goals of the American educational system. However, according to data from the National Assessment of Educational Progress (NAEP), there is great disparity between the levels of math literacy for students with disabilities when compared to the results for students without disabilities. Research compiled by the National Science Foundation further shows that students with disabilities exit high school with significantly fewer course credits in mathematics and science subjects than students without disabilities. Overall, students with disabilities are much less likely to graduate from high school and enroll in postsecondary education--and among those students with disabilities who do graduate, they generally take fewer science and math courses, had lower grades, and had lower achievement scores than their peers without disabilities (National Science Foundation, 2003). There are undoubtedly many factors at work which have a connection to the poor math performance of students with disabilities. A fundamental contributing factor is that virtually all mainstream math instructional content is not designed to be utilized with the assistive technology products that many students with disabilities use, and is thus not accessible. This is especially true of print classroom textbooks, which are commonly used to determine the instructional math program for students in most school settings. 75% to 90% of all classroom instruction is based on textbooks, and, in most cases, those books define the scope and sequence of the material being taught (Tyson & Woodward, 1989). This is also the case with math instruction, where 80% to 90% percent of grades 4 - 12 math and science classrooms use textbooks (Hudson & McMahon, 2002). 7


Of late, many in the education community have turned to the use of digital texts which can be transformed with assistive technologies into an audible version of the textbook using synthetic speech on a computer. Such conversions to digital formats have been a boon to providing equitable instructional access for students with disabilities who use these technologies. "Technology allows print textbooks to be made accessible to students with disabilities through conversion to digital form. The same material in digital form offers many options for students with disabilities. It can, for instance, be read aloud by a computer or screen reader, or printed on a Braille printer. The power of future curriculum will be in these alternative digital formats." (Stahl and Aronica, 2002) The use of digital texts, however, has been largely focused on providing access to standard literary materials, rather than to math content. Higher level math access with assistive technologies is particularly problematic, due to the fact that common scanning and optical character recognition (OCR) technologies used to convert print materials to digital form cannot process complex math symbols, and publisher created digital resources commonly use inaccessible graphical images of math equations. Unfortunately, this lack of available accessible digital content in areas of math instruction is perhaps even more problematic for students whose disabilities affect reading comprehension. This is because of the additional mental processing that is required to interpret math expressions compared to literary content. Such an understanding may be supported by the fact that more than 60% of students with learning disabilities which affect reading comprehension, for example, have been shown to possess significant disabilities in mathematics (Light & DeFries, 1995). A number of studies have found that students with learning disabilities experience more significant difficulties in acquiring math skills than do their peers without disabilities (Miller & Mercer, 1997). Research has also shown that students with language deficits react to math problems on the page as signals to do something, rather than as meaningful sentences that need to be read for understanding (Garnett, 1998). In particular, this research points out that many students with learning disabilities have a tendency to avoid verbalizing in math activities. Such findings tend to reinforce the concept that LD students with math deficits are seemingly unable to self-verbalize math equations. Computerized reading of math equations could therefore aid students by both reinforcing self-verbalization skills as well as providing access to content for students whose disabilities prevent effective self-verbalization of math equations. Student access issues further accumulate with increasingly difficult mathematics as students attempt to understand the meaning and syntax of mathematical expressions that occur in the study of higher math subjects such as algebra and calculus. Such mathematical disciplines incorporate a distinct symbolic language which students must learn to recognize and decode as an essential task to developing math literacy skills. The capability to hear math equations properly decoded and verbalized could therefore be an important accessibility component for students with learning disabilities. Similarly, for students who are blind, the ability to have math equations unambiguously spoken by computer technology, or able to be used with refreshable braille displays, can be a vital accessibility technique.

8


THE NEED FOR UNIVERSAL DESIGN FOR MATH LEARNING Standard print textbooks and other types of commonly used instructional materials are inaccessible to a large percentage of students with disabilities and usually require transformation into alternative formats, such as recordings, braille or accessible digital formats to provide access to students with various print disabilities (Stahl, 2004). Math textbooks and other instructional materials will provide much greater accessibility for students with visual or learning disabilities when they are made available in a universally designed accessible digital format. Such a universally designed digital format for math content can be achieved by using Mathematical Markup Language (MathML). MathML is an open industry standard first adopted by the World Wide Web Consortium (W3C) in 1998. MathML is an XML-based application for describing mathematical notation and capturing both its structure and content. Using MathML enables mathematics to be served, received, and processed in digital environments such as the World Wide Web, just as Hyper Text Markup Language (HTML) has enabled this functionality for literary text. Most importantly for this discussion, using MathML provides for a standard approach to content tagging and information structure which can make mathematical information available to assistive technology in a way that is comparable to standard access by students without disabilities. MathML provides the technology foundation for accessible math—as opposed to just graphical images that are the current norm. Using MathML will allow assistive technologies to provide built-in alternate access avenues, such as using synthetic speech to read math equations out loud, or providing for seamless text enlargement, braille support, and providing a means for students to navigate both visually and aurally through complex math formulas and highlight expressions as they are read. Until very recently, speech synthesis technology has been unable to process complex math equations, and virtually all digital math content has been produced using graphical images which are inherently inaccessible to either text reader or screen reader assistive technologies. Design Science, Inc., a developer of mainstream math publishing technology, has been engaged in research and development efforts since 2003 supported in part by the National Science Foundation (SBIR Grant No. 0340439) to make math expressions created with MathML seamlessly accessible to people with visual or learning disabilities. One of the fundamental principles of this work has been to provide integrated access to mathematical content through users’ existing screen readers or other assistive technology. The advantage of this approach to math accessibility is that it allows materials containing math to be read with standard browsers and familiar assistive technology devices instead of depending upon a stand-alone proprietary application. MathPlayer, made by Design Science, provides the state-of-the-art in audio rendering of mathematical expressions, navigation of mathematical expressions with audio feedback, and audio rendering synchronized with highlighting of the expression being spoken. Commonly used assistive technologies, such as JAWS, Window-Eyes, HAL, MAGic, Read&Write and BrowseAloud, can take advantage of MathPlayer's unambiguous speech access in a transparent

9

First International Conference on Technology-based Learning with Disability manner which works in concert with the user's usual technology environment. For assistive technologies that support word highlighting, MathPlayer also integrates math highlighting so that both the words and the math expressions are highlighted as they are spoken. These accessibility provisions, however, all depend upon the digital source math content being made available in MathML. CONNECTING MATHML WITH NIMAS Under the provisions of the Federal Individuals with Disabilities Education Act (IDEA) of 2004 and its implementing regulations, publishers are now providing textbook content using the National Instructional Materials Accessibility Standard (NIMAS). NIMAS-compliant textbook files are sent by publishers to the National Instructional Material Accessibility Center (NIMAC), which in turn provides these files to third-party accessibility entities who distribute student-ready versions to students with print disabilities. These provisions can contribute greatly to schools meeting the No Child Left Behind (NCLB) requirements for Adequate Yearly Progress (AYP) by providing curricular content in usable form at the same time that opportunity to learn exists for all other students, which is a prerequisite for participation in standards-based reform and accountability (Elmore, R.F., 1995; Guiton, G. & Oakes, J., 1995). The advent of the National Instructional Materials Accessibility Standard offers students with print disabilities significant newfound opportunity for access to the general curriculum and learning by the flexibility of use of digital content. In his recent testimony to the NCLB Commission, Dr. David Rose, CEO of the Center for Applied Special Technology (CAST), stated “An important first step in ensuring this flexibility has recently been signed into federal law—the National Instructional Materials Accessibility Standard (NIMAS). This standard requires that publishers of print materials, e.g. textbooks, must provide flexible alternatives— digital versions—for students with “print disabilities.” These alternatives provide alternate paths to the same high standards for students who cannot see or successfully decode traditional textbooks”. (Rose, 2006) The NIMAS specification is essentially a subset of a larger industry standard created for the production of digital talking books, called the Digital Accessible Information System (DAISY). The federal regulations defining NIMAS identifies which DAISY tags are mandatory, and which are optional (US Department of Education, 34 CFR Part 300). Although the original DAISY specification upon which the 2006 NIMAS was based did not specify a way to integrate MathML into digital publisher files, this problem has now been solved. The DAISY specification does define how Modular Extensions can be added to the standard to deal with non-literary materials, and therefore the DAISY Consortium has recently developed a solution for including mathematics using a modular MathML extension, enabling full support for accessible mathematics in the DAISY/NISO Standard. The publication of the Mathematics Modular Extension is thus crucial to integrating accessible mathematics via MathML into DAISY and NIMAS-compliant books. Chuck Hitchcock, Director of the NIMAS Technical Assistance Center at the Center for Applied Special Technology (CAST) indicates that the publication of the DAISY Modular Extension for math is an important advance toward the universal design of math instructional content. "Now that DAISY has integrated a MathML vocabulary into its specification, publishers creating NIMAS-compliant files as part of federal IDEA requirements

10

First International Conference on Technology-based Learning with Disability will soon be able to support a much greater level of accessibility and educational efficacy for elementary and secondary math textbooks" (DAISY, 2007). CONCLUSION: BRIDGING THE TECHNOLOGY AND POLICY DIVIDE In conclusion, we should return to our original premise, that making math accessible is as much a public policy issue as it is a technological one, and that the concept of what is required under law (public policy) continues to expand as the technological issues of effective access are resolved. Now that the technological issues of accessible math have been resolved--by virtue of MathML and its adoption within NIMAS--it is essential that disability advocates and educators push for better public policy to support the availability of accessible math in the classroom. There are a number of policy vehicles available to push this forward: 1) Updating of Federal NIMAS specifications As previously mentioned the current NIMAS specification is tied to DAISY, but does not have an explicit reference to the new DAISY MathML modular extension. Without such an explicit linkage, one can only assume that the MathML extension will be left as an optional requirement, leaving the actual enforcement of the requirement to state by state interpretations. The NIMAS Development Center at CAST is tasked under Federal regulations to update the NIMAS specification, so this will be an important mandate for CAST to propel through the regulatory change process as soon as possible. 2) State Implementation of NIMAS and MathML States are the primary implementing entities under the Federal NIMAS requirements. States--as well as school districts in states not having statewide textbook adoptions--have the autonomy to enforce publisher requirements on their own, even without a specific Federal mandate. States should use this ability to push their accessibility agenda forward and require that publishers utilize MathML when preparing math textbooks for submission to the NIMAC. 3) State and District purchasing requirements States and school districts have federal obligations under Section 504, the Americans with Disabilities Act, and the IDEA to make their instructional content accessible to students with disabilities. Since the advent of MathML has made accessible math a reality, educational entities must move forward to ensure that math instructional content created with MathML will be available to their students who use assistive technologies. Beyond placing MathML requirements in textbook adoptions and contracts, states should also put these requirements in Requests for Proposals (RFPs) for statewide assessment, so that students can utilize accessible math in online assessments as well. Furthermore, when states and districts negotiate major site license relationships with assistive technology vendors, they should ensure that these contracts require the vendor to support MathML technology. Including MathML requirements in procurement policies for instructional software is yet another vehicle for furthering the availability of accessible math technologies for all students. These are some of the most direct methods for securing effective public policy to support accessible math. Such actions will help to ensure that educational content in MathML will

11

First International Conference on Technology-based Learning with Disability become available to students as quickly as possible, and as broadly as feasible. Together, we can move forward toward an inclusive universal design for learning environment for the study of math. REFERENCES DAISY Consortium Press Release: Mathematics Now Added to the DAISY Standard. March 22, 2007. Accessed at: http://www.daisy.org/news/news_detail.asp?NewsId=296 Elmore, R. F. (1995). Structural reform in educational practice. Educational Researcher, 24, 2326 Garnett, Kate. (1998). Math Learning Disabilities. Accessed at http://www.ldonline.org/article/5896 Guiton, G. & Oakes, J. (1995). Opportunity to learn and conceptions of educational equality. Educational Evaluation and Policy Analysis, 17(3) 323-336 Hudson, S.B., McMahon, K.C. & Overstreet, C.M. (2002). The 2000 National Survey of Science and Mathematics Education: Compendium of Tables Authors. Horizon Research. Light, G. J., & DeFries, J. C. (1995). Comorbidity for reading and mathematics disabilities: Genetic and environmental etiologies, Journal of Learning Disabilities, 28, 96-106 Miller, S., & Mercer, C. (1997). Educational Aspects of Mathematics Disabilities. Journal of Learning Disabilities, 30 (1), 47-56. National Science Foundation, Division of Science Resources Statistics (2003). Women, Minorities, and Persons With Disabilities in Science and Engineering: 2002, NSF 03-312. Rose, D., Testimony to NCLB Federal Commission August, 2, 2006, Aspen Institute in Washington, DC Stahl, S & Aronica, M. (2002). Digital Text in the Classroom Journal of Special Education Technology, 17 (2), spring 2002 (accessed from http://jset.unlv.edu/17.2T/tasseds/rose.html) Stahl, S. (2004). The promise of accessible textbooks: increased achievement for all students. Wakefield, MA: National Center on Accessing the General Curriculum. Retrieved 05-26-06 (accessed from http://www.cast.org/publications/ncac/ncac_accessible.html) Tyson, H., & Woodward, A. (1989). Why students aren't learning very much from textbooks. Educational Leadership, 47 (3), 14-17. US Department of Education. 34 CFR Part 300. Accessed at: http://nimas.cast.org/about/proposal/register-v1_1.html

12


Novel Approaches to Deaf Education Nicoletta Adamo-Villani Purdue University West Lafayette, IN Ronnie Wilbur Purdue University West Lafayette, IN ABSTRACT In this paper, we describe the development of two novel approaches to teaching math and science concepts to deaf children using 3D animated interactive software. One approach, MathsignerTM, is non-immersive and the other, SMILETM, is a virtual reality immersive environment. The content is curriculum-based, and the animated signing characters are constructed with state-ofthe art technology and design. We report preliminary findings. INTRODUCTION This paper presents two novel approaches to deaf education using 3-D animation technology, one non-immersive and one immersive. Both approaches described here are unique because they: (1) use advanced technology to teach mathematics to K-6 deaf students who know American Sign Language (ASL); (2) provide equal access and opportunities by overcoming known deficiencies in science, technology, engineering, and math (STEM) education as reflected in the underrepresentation of deaf people in fields requiring STEM skills; and (3) provide a model for teaching technology in general that can contribute to improving deaf education around the globe. Our expertise in language problems of deaf children and linguistic research on ASL structure enables these programs to be appropriate in both English and ASL. MATHSIGNER™ – A NON-IMMERSIVE GAME FOR STANDARD COMPUTERS Mathsigner is a 3D animation ASL-based interactive software package which contains sets of activities, with implementation guidelines, designed to teach K-6 math concepts, signs, and corresponding English terminology to deaf children, their parents, and teachers. Mathsigner is being developed using cutting-edge 3D animation technology. Computer generated and controlled animation presents many advantages over other technologies, including: (a) User control of appearance - orientation of the image (rotation and point of view control); location of the image relative to background; size of image; zoom. (b) Quality of the image - no distracting details as in photos and films; texture and transparency control. (c) User control of the speed of motion. (d) User programmability for: generating infinite number of

13

First International Conference on Technology-based Learning with Disability drills; unlimited text encoding; real time translation; limitless combinations of signs. Manual signs and facial expressions can be combined in any manner under program control. (e) Whole sentences can be linked together smoothly, without abrupt jumps or collisions between successive signs as would happen combining video clips. (f) Very low bandwidth. The programs controlling animations can be stored and transmitted using only a few percent of the bandwidth required for comparable video. (g) Character control. Animated signs can be easily applied to other characters, including different ages and ethnicity as well as cartoon characters. Innovations in Mathsigner Compared to Other 3D Animation-Based Signing Products Accuracy and realism of the signs: We use state-of-the-art optical motion capture system to record the signs directly from a fluent signer, which are captured by 6 cameras and applied to 3D characters in real time for immediate feedback and editing on readability and realism. Smooth transitions between individual signs: The authors have filed a patent for a technique that allows for real-time blending of individual animation segments which yields smooth signed sentences from sequences of single signs. In other 3D signing programs, the transitions between signs are implemented with cut-and-paste methods or simple linear interpolation; the result is unrealistic in-between movements. A detailed description of the blending technique can be found in (Adamo-Villani, Doublestein & Martin, 2005) High quality character appearance - organic deformations during signing motion: We use stateof-the-art modeling and rigging techniques to model 3D signers (realistic and fantasy) as seamless polygonal models, a major improvement over the appearance of existing segmented or partially segmented signing avatars, which do not change realistically as they move. Natural facial expressions: One of the authors has developed a parameterized graphical facial model with a set of 26 parameters each controlled by a letter on the keyboard (US patent pending) (Adamo-Villani & Beni, 2004). The method allows encoding in real-time of significant facial expressions with accuracy and realism. Such facial modeling will represent improvement over existing avatars whose facial expressions are mechanical and limited to a small set. Ready-to-use software for K-6 math education: Our software is not just animated signing, but an integrated package with math activities sorted by grade level concepts and difficulties. The software can be delivered via web or CD-ROM; no special system requirements or further programming is required to run the program. No other interactive, 3D animation-based software exists for math education of the Deaf. The Prototype We have a fully functional prototype that teaches grades K-3 math concepts and related ASL signs. The math content is based on standard, published elementary school math curriculum and has been developed with feedback from the teachers at the

14

First International Conference on Technology-based Learning with Disability Indiana School for the Deaf (ISD). The current prototype contains two programs, one aimed at deaf children and the other aimed at hearing parents. Each has two modes of operation - a learning mode and a practice/drill mode. The two modes of usage are characterized by different color schemes (yellow for learning and orange for testing). The screen layout (shown in Fig.1) consists of two frames. The frame on the left is used to select the grade (K-1, 2 or 3) or the type of activity. The frame on the right shows the 3D signer. The upper area on the left (in green) gives textual feedback as appropriate; the bottom area shows the navigational buttons. The frame on the right contains a white text box below the signer, to show the answer (in mathematical symbols) to the current problem. Below this, there is a camera icon and an arrow. The arrow (slider) is used to control the speed of signing; the camera button opens a menu to zoom in/out on the 3D signer, change the point of view and pan to the left or to the right within the 3D signer window. A demo of the prototype learning tool is available at http://www2.tech.purdue.edu/cgt/I3/. Design Improvements The prototype has been evaluated throughout its development by ASL signers, faculty, and students knowledgeable in sign language and deaf education. These informal evaluations have produced key findings that are currently being used to modify and improve the design of the application. To summarize, the evaluations have produced results at three levels: 1. Recommendations for improved interaction 2. Recommendations for enhancement of overall appeal of the application 3. Suggestions for improved character design To improve interaction with the application, the screen layout was changed from two to three panels (Fig. 2). Now the tasks of learning and testing are clearly separated. The signer is placed in the middle and, as a result, the user can fully attend to the center and use peripheral vision while moving the cursor over the buttons. In the prototype, the user had to look at the left side to place the cursor on a button, and then at the right side to understand what the button meant. This continuous shift in gaze direction was tiring and difficult to maintain for a long period of time. Several changes were made to the look of the interface to make it more appealing to the target age group. The icons were redesigned to be more age appropriate and visual distraction was reduced. The screen background now changes when different 3D signers are selected. Fig. 3 shows the screen design that appears when the “space signer” is selected. Significant changes were made to the original “bunny signer”. The new “bunny” (represented in Fig. 2) has a more human-like anatomy and therefore can sign more clearly. In addition, new characters are being developed; one of them is represented in Fig. 3.

15


Evaluation Plan Evaluation of Mathsigner is done with ISD. The formative evaluation focuses on design features and quality of the signing motion. Program success is determined by: (1) deaf children's reactions (willingness to use, time on task), and (2) teachers’/parents’ feedback on the degree to which the program help to meet teachers/parents’ math goals at each grade level. The animated signing is evaluated by experts who rate it (scale of 1 to 5) on several factors: realism of signing motion, readability, fluidity of transitions between signs, motion timing, sign placement in relation to body, and sign accuracy. So far, the feedback on signing motion has been very positive, especially readability, fluidity, and timing. Placement and accuracy feedback led to the character redesign, including clearly delineated neck and a longer torso to permit greater separation on the vertical axis of the locations of sign formation with respect to the body. Summative evaluation will start in Spring 2008 and will test the efficacy of Mathsigner with three main questions: (1) Does Mathsigner lead to a learning effect? We compare scores on the SAT (Stanford)-HI (hearing impaired) and SESAT (Stanford Early School Achievement Test for younger children) mathematics subtests in the form: pre-treatment, treatment (use of software), post-treatment. (2) Is learning through Mathsigner more efficient than standard techniques? We compare our student scores with historical norms from the SAT-HI using both the national hearing-impaired norms (available from Gallaudet Research Institute) and the ISD norms. (3) What factors affect learning through Mathsigner? Our hypothesis is that learning gains are correlated with learner effort. We use several ‘effort’ measures including: (1) time spent logged on to Mathsigner; (2) number of items completed; and (3) number of attempts at each item. SMILE (SCIENCE AND MATH IN AN IMMERSIVE LEARNING ENVIRONMENT) – AN IMMERSIVE GAME FOR STATIONARY AND PORTABLE VR SYSTEMS SMILE is an immersive Virtual Learning Environment (VLE) in which deaf and hearing children ages 5-10 learn STEM concepts and ASL terminology through user interaction with fantasy 3D characters that communicate in ASL and spoken English. Background Research in VR and education is a young field which has recently shown considerable growth. Youngblut reports over forty VR-based learning applications (Youngblut, 1997) and Roussou describes about 10 VLE designed for informal education (Roussou, 2004). Although the benefits of VR experiences need to be more comprehensively defined, studies show that VR can provide a more effective learning tool than traditional classrooms, students enjoy working with virtual worlds, and the experience is highly motivating (Youngblut, 1997). Research also shows that VR is particularly suitable to STEM education because of its ability to ‘bridge the gap between the concrete world of nature and the abstract world of concepts and models’, making it a valuable alternative to the conventional study of math and science which requires students to develop understandings based on textual descriptions and 2D representations (Johnson et al. 2002).

16


Regarding disabilities education, VR has advantages over other teaching technologies because it can provide for the learning requirements of students with disabilities (Darrow, 1995). Some of the most commonly encountered needs include: access to safe and barrier-free scenarios for daily living tasks; self-pacing; repetition; control over environment; ability to see or feel items and processes in concrete terms (difficulty with abstract concepts); and motivation. Roussou suggests that there are many compelling reasons for believing that VLE provide effective teaching tools for children’s conceptual learning (Roussou et al.1999). However, due to the use of high-end expensive equipment and non-standard application development, the majority of existing VLE for children is limited to academic and research environments and institutions of informal education, such as museums. One notable example of VLE is the NICE project (Roussou et al., 1999) designed for display in the CAVE VR system (Cruz-Neira, Sandin, & DeFanti, 1993). NICE is an immersive, multi-user VLE in which children learn basic biological concepts while constructing, cultivating, and tending a virtual garden. VREAL (Virtual Reality Education for Assisted Living) project (Edge, 2001) is, to date, the only VLE for deaf children. VREAL is an immersive virtual environment in which deaf students learn basic life skills, language arts, and mathematics. Five US Deaf schools used the program in 2004 and assessment studies showed a student test score improvement by an average of 35%. SMILE follows the trail pioneered by projects such as VREAL and NICE, but makes unique contributions to this area. (1) SMILE is the first bilingual immersive VLE featuring interactive 3D animated characters that respond to the user’s input in English and ASL. (2) It includes significantly improved seamless characters compared to existing 3D animated signing, i.e., Signing Avatar (Vcom3D, 2004), with fluidity of signing motion and realism of skin deformations. (3) Its content is designed by a team of experts including specialists in VR application development, ASL and Deaf education, STEM education, graphic design, animation, and game design. Roussou and Barab (Roussou et al.1999; Barab et al., 2005) argue that highend technological innovations are often associated with disappointing content. SMILE attempts to provide an ideal combination of technological innovation and educational content by presenting an emotionally appealing visual design, an engaging metagame strategy that establishes a meaningful context for participation, and goal-oriented activities that are grounded in research on effective pedagogy. (4) SMILE is designed for formal STEM education and will be available for use by elementary schools and deaf education programs throughout the US. Students will interact using a relatively inexpensive projection-based portable system that eliminates the cumbersome HMD unit, while maintaining the feeling of immersiveness. Development of SMILE SMILE is an interactive virtual world containing an imaginary town of fantasy 3D avatars that communicate with the user in written and spoken English, and ASL. The user can explore the town, enter buildings, select and manipulate objects, construct new objects, and interact with the characters. In each building the use learns specific STEM concepts by performing hands-on activities developed with elementary school educators (including deaf educators), and in

17

First International Conference on Technology-based Learning with Disability alignment with standard STEM curriculum. SMILE has an overall story which is introduced through a cutout-style 2D animation at the beginning of the game (fig. 4 shows a frame). The story includes an overarching goal (restore the lost willingness to smile in the city of ‘Smileville’) which creates a boundary condition that unites all the individual game tasks. Each activity is in the form of a ‘good deed’ whose objective is to make a ‘Smileville’ character smile again by giving him/her a meaningful new object. The ability to construct the object is dependent on the acquisition of STEM skills, and related ASL signs. All game activities are carried out in a cartoon-like virtual world designed to be appealing to the target age group. Key design features include basic geometric shapes with round edges, vibrant and varied colors, and a bright lighting setup with limited shading and soft shadows. The choice of the color and lighting schemes was based on research studies on the impact of color and light on learning (Duke, 1998) (Engelbrecht, 2003), and on the association between colors and children’s emotions (Boyatzis & Varghese, 1994). The visual and game designs of SMILE are described in detail in (AdamoVillani & Wright, 2007). Fig. 5 shows the exterior and interior of the bakery building, and one of the 3D characters.

The character design is very stylized and consistent with the environment visual style. All characters are modeled as continuous polygon meshes with a poly-count less than 6000 polygons per avatar. A low polygon count maintains a high frame rate and real-time interaction. To realize high visual quality with a limited number of polygons, the 3D surfaces have been optimized by concentrating the polygons in areas where detail is needed the most: the hands, the face, and the parts that bend and twist (i.e. elbows, shoulders, wrists, and waist). With such distribution of detail it is possible to represent realistic hand/face configurations and organic deformations of the skin during motion. Character bodies are set up for animation with a skeletal structure closely resembling a real human. The face is rigged with 20 to 30 joint deformers positioned so that they deform the digital face along the same lines pulled and stretched by the muscles of a real face. For fluidity and realism, the signing uses the same techniques as the Mathsigner project. Technical implementation SMILE™ can be displayed on different systems: (1) stationary 4-wall projection devices (i.e. the Fakespace FLEX); (2) single screen portable projection systems; (3) 18

First International Conference on Technology-based Learning with Disability Fish Tank VR systems, and (4) standard desktop computers. The application could also be modified to be viewed through a head mounted display unit. The development of SMILE is described in detail in (Adamo-Villani, Carpenter & Arns, 2006) SMILE™ in the FLEX: The student views the application through a pair of light-weight LCD active stereoscopic glasses projected onto the immersive, four screen display (see fig. 6), which provides the user with virtual environment images projected to the front, side, and floor screens. The user wears an InterSense head tracker to determine the position and orientation of the eyes; this information redraws the environment based on the user’s perspective, as the direction of gaze changes. The user travels through the environment using an Intersense 6 DOF wand or a Cobult Flux dance platform. Objects can be selected and manipulated with the wand, or with a simple gesture control system (a pair of Fakespace Lab’s Pinch Gloves coupled with an Intersense wrist tracker). The gesture control system allows for input of ASL signs (numbers 0-20), and for travel through the environment. SMILE™ on portable systems: Because SMILE™ is designed primarily for display in a fourwall projection system, certain objects in the scene exist in the user’s peripheral vision and on the floor of the scene. We have developed a portable version which eliminates unnecessary information from the sides of the environment and moves the important features to the front of the user’s view. This transition from 4-wall display to a single monitor has been accomplished by editing the VRJuggler configuration files. Adding additional devices, such as LCD shutter glasses for CRT monitors or desktop tracking systems, require nothing more than the installation of new device drivers and the creation of new configuration files. SMILE™ has been tested on the following portable systems: (1) a projection-based immersive system consisting of a screen and frame, a high-end laptop, two commodity projectors, a pair of polarizing filters, and inexpensive polarized glasses. (2) a Fish Tank VR system consisting of a DellE520 desktop PC, a CRT monitor, an Essential Reality P5 glove with 6 degrees of tracking and bend sensors for each finger, a pair of eDimensional wireless 3D glasses and an Intersense 3DOF PC head tracker. (3) a standard, non immersive desktop computer system. The application is designed mainly for use with the Intersense IS-900 system.When a tracking system is not available, input can be accomplished via mouse and keyboard. Portable demos of SMILE are available for download at: http://www2.tech.purdue.edu/cg/i3/smile/demos Evaluation of SMILE The evaluation of SMILE includes three forms: expert panel-based, formative, and summative. The expert panel-based and formative evaluations focus on the usability and fun, visual representation quality, and signing motion quality, and are repeated throughout the development of SMILE to identify recommendations for design improvement..The panel consists of experts in VR application development, 3D modeling and animation, and American Sign Language. Each evaluator is asked to perform an analytical assessment in his/her area of expertise. The experts in VR application development have so far assessed the usability of the program by determining what design guidelines it violates and supports. Clear heuristics for the ideal design of VE to guide such evaluation do not exist yet; guidelines used by the experts were derived from work by (Nielsen & Molich, 1990; Nielsen, 1994; Gabbard, 1998).

19

First International Conference on Technology-based Learning with Disability The 3D modeling and animation experts have been given questionnaires focusing on the visual representation of the virtual world; the experts in ASL have been given questionnaires on the quality of the signing motion. To date, several usability problems have been uncovered and solved; all elements of the visual representation and signing motion have been given high scores by the experts and, therefore, recommendations for improvement have not been necessary. Three formative evaluations with target users have been administered so far (next section). Summative evaluation assesses learning. Such evaluation with kindergarten and elementary school aged deaf and hearing children will be done in collaboration with the Indiana School for the Deaf (ISD) in Indianapolis, and with two elementary schools in West Lafayette, IN. Summary of Initial Findings The procedure and evaluation instrument used in the first three formative evaluations are described in detail in (Adamo-Villani & Wright, 2007) and (Adamo-Villani & Jones, 2007). Overall, children enjoyed playing the game and found the environment and characters fun and appealing. Although they had high expectations, the reported experience surpassed them. SMILE was perceived more fun and easier to use than expected, and slightly more challenging. The ‘Again-and-Again’ table (Read, Macfarlane & Casey, 2002) revealed that the activities the children most enjoyed were the construction of the new objects (i.e. the cake), watching the mysterious machines (such as the animated baker’s machine), traveling through Smileville, and playing the entire game. Observation and think aloud protocol showed that other activities the participants found ‘very fun’ were ‘walking through objects’, ’throwing objects’ ,‘opening doors’, and ‘watching things that move’. As far as usability, children did not appear to have major difficulties with travel, selection, and manipulation tasks. We noticed a few signs of frustration and comments such as ‘some of the objects are really hard to pick up’ and ‘some of the text is hard to read’. Two subjects showed discomfort (dizziness and eye strain) with the head tracker and glasses and stopped interacting with the application after approximately 10 minutes. The main problem was the size of the 3D shutter glasses. Children kept losing the goggles during interaction and were constantly adjusting them on their noses. We are researching different solutions such as customized 3D glasses for children, coupling the goggles with a head band, or using a 3D monitor that does not require glasses (for the Fish tank VR system). As for engagement, the majority of the students appeared to be very focused on the tasks. Positive comments included: ‘this is awesome because you feel like you are really in a bakery’; ‘….this game is more exciting than a video game because you don’t see anything around you… and you are really inside the building putting a cake in the oven’. Many positive signs were observed such as laughing, smiling, bouncing in excitement, and ‘wow’ sounds. ACKNOWLEDGEMENTS This research was supported in part by National Science Foundation HRD-0622900. We appreciate the assistance of the Indiana School for the Deaf and the Envision Center for Data Perceptualization at Purdue University.

20


REFERENCES Adamo-Villani, N., & Beni, G. (2004). Keyboard encoding of facial expressions. IEEE Proceedings of 8th International Conference on Information Visualization-HCI symposium (IV04), 14-16 July 2004, London, 324-328. Adamo-Villani, N., Carpenter, E., & Arns, L. (2006). An immersive virtual environment for learning sign language mathematics. ACM Proceedings of Siggraph 2006 - Educators, Boston. Adamo-Villani, N., Doublestein, J., & Martin, Z. (2005). Sign language for K-8 mathematics by 3D interactive animation. Journal of Educational Technology Systems, 33 (3), 243-259. Adamo-Villani, N., & Jones, D. (2007). Travel in SMILE: a study of two immersive motion control techniques. Proceedings of IADIS International Conference on Computer Graphics and Visualization 2007, Lisbon (accepted) Adamo-Villani, N., & Wright, K. (2007). SMILE: an immersive learning game for deaf and hearing children. ACM Proceedings of Siggraph 2007- Educators, San Diego (accepted). Barab, S., Thomas, M., Dodge, T., Carteaux, R., & Tuzun, H. (2005). Making learning fun: Quest Atlantis, a game without guns. ETR&D, 53, 1, 86-107. Boyatzis, C.J., & Varghese, R. (1994). Children’s emotional associations with colors. Journal of Genetic Psychology, 155, 1, 77-85. Cruz-Neira, C., Sandin, D. J., & DeFanti, T. A. (1993). Surround-screen projection-based virtual reality: The design and implementation of the CAVE. Proceedings of ACM SIGGRAPH '93, Anaheim, CA. Darrow, M.S. (1995). Virtual reality's increasing potential for meeting needs of persons with disabilities: What about cognitive impairments? Proc. of the Annual International Conference on Virtual Reality and Disabilities, Northridge, CA: California State Center on Disabilities. Duke, D. L. (1998). Does it matter where our children learn? White paper for the National Academy of Sciences and the National Academy of Engineering. Charlottesville: University of Virginia. Engelbrecht, K. (2003). The impact of color on learning. NeoCON2003. http://www.coe.uga.edu/sdpl/articleoftheweek/colorPW.pdf Gabbard, J.L. (1998). A taxonomy of usability characteristics in virtual environments. Master’s thesis, Dept. of Computer Science and Applications, Virginia Polytechnic Institute and State University. Johnson, A., Moher, T., Choo, Y., Lin, Y.J., & Kim, J. (2002). Augmenting elementary school education with VR. IEEE Computer Graphics and Applications, 22, 2, 6-9. Nielsen, J., & Molich, R. (1990). Heuristic evaluation of user interfaces. Proc. of ACM CHI'90 Conference, Seattle, 249-256. Nielsen, J. (1994). Heuristic evaluation. In Nielsen, J., and Mack, R.L. (Eds.), Usability Inspection Methods. John Wiley & Sons, New York, NY. Read, J. C., Macfarlane, S. J., & Casey, C. (2002). Endurability, engagement and expectations: Measuring children's fun. Interaction Design and Children, Eindhoven: Shaker Publishing Roussou, M. (2004). Learning by doing and learning through play: an exploration of interactivity in virtual environments for children. ACM Computers in Entertainment, 2, 1, 1-23. Roussou, M., Johnson, A., Moher, T., Leigh, J., Vasilakis, C., & Barnes, C. (1999). Learning and building together in an immersive virtual world, Presence, 8, 3, 247-263. Youngblut, C. (1997). Educational uses of virtual reality technology. VR in the Schools - coe.ecu.edu, 3, 1.

21


AutOMathic Blocks – An Automated Systems to Teach Math to K-12 Children with Severe Visual Impairment Allowing Both Physical and Haptic Interaction with An Automated Tutor1 Arthur I. Karshmer, Priyanka Daultani and Michael McCaffrey University of San Francisco, USA Abdel Ejnioui University of South Florida, USA Roope Raisamo University of Tampere, Finland Abstract Learning mathematics has always been a daunting task for the visually impaired student. For the most part, a task that restricted their entry into careers based on the reading and writing of mathematical equations. There have been some notable exceptions, but for the most part education in the domains of math, physics, computer science and engineering have been beyond the grasp of blind students. In the current work, we a new system being developed to help young blind students learn both arithmetic and basic algebra. Introduction Imagine yourself a second or third grader learning complex addition. Your are given the following exercise to solve.

7 4 8 + 1 2 Not a terribly difficult problem, but it does require the concept of “carry” to arrive at the proper solution. As you begin to write the solution, the teacher notices that you have written 50

7 4 8 + 1 2 5 0 as a partial answer to the problem, and recognizes that you are on the way to an incorrect solution. You have neglected to consider the carry from the least significant column. If you were 1

The work reported here was supported by a grant from the University Of San Francisco’s Jesuit Foundation

22

First International Conference on Technology-based Learning with Disability a sighted student, after some mentoring by your teacher, you would simply erase the 5, replace it with a 6 and carry on with your work. If, on the other hand, you are a visually impaired student, you would face several problems in solving this exercise. A brief list of these issues would include, 1. The printed presentation would be of no use to you. 2. If you are at least, a minimally proficient Braille user, there is a high probability that your teacher would not be able to help, as he/she would most likely not be proficient in Braille. The problem can be mitigated through the use of a simple, but efficient tool – Braille math blocks. Using this simple, but effective tool, your teacher needs no knowledge of Braille, and can follow your progress in the problem solving process (See figure 1). If the arithmetic is simple, the teacher is facile in arithmetic, the teacher has the proper training in special education and has the extra time to spend with you at school and at home, the problem solved! Well, not likely. Arithmetic rapidly gets more complex as the student learns subtraction, multiplication and division and with these complexities comes more problems for the visually impaired student. The next steps include simple algebra and geometry - and even less human support in the learning process. Now, most parents are effectively out of the teaching and mentoring loop and the paucity of special education math teachers becomes a bigger and bigger problem.

Figure 1. Braille Math Blocks

The AutOMathic block system would have tracked the student through the simple example above giving the following advice at the same point, or earlier, than the teacher or parent would have intervened. You seem to have forgotten the carry from the previous column.

Figure 2. A sample of feedback from the system which is delivered in an audible form.

23


All system feedback comes from a computer working together with a specialized version of the Braille math blocks and is rendered via speech or refreshable Braille, or both. Multiple, national spoken and Braille-based information will be supported by the system using tools already developed. The AutOMathic system is not restricted to simple arithmetic, or actual physical blocks. In parallel with the development of the physical block system, a virtual block system based on haptic feedback will also be developed. Both will include the same teaching and mentoring tools. From simple addition to algebra, the AutOMathic block system will provide the visually impaired student their own personal teacher / mentor, and the cost of the final system will be low enough to be affordable by schools and students. While the current example seems quite simple, it is only the forerunner of much more complex arithmetic and mathematics. And, without the basics, later materials will become difficult or impossible to learn. Foundational material is critical to continued learning of math and then to other math based disciplines such as physics, chemistry, engineering, and the list goes on. The AutOMathic blocks system is designed to take visually impaired students from basic arithmetic through algebra concepts. Key elements in building a foundation to enter STEM careers. A Systemic Problem Education for the visually impaired continues to be a matter of concern. The problem spans almost all aspects of education, but is most notable in the study of STEM (Science, Technology, Engineering and Mathematics) disciplines. While considerable work has been done in recent years [Karshmer – 2005, Edwards – 2006] to solve parts of this problem, these efforts have been aimed at specific subsets of the overall issue – but not at the early years of the educational experience. A most critical period. Two European projects of note, the MATHS [Gill - 2006] project of several years past and the current LAMBDA project [Edwards - 2006] have different approaches to solving the problem. The MATHS project depended on special purpose hardware that never materialized and the LAMBDA project proposes the development of an entirely new representation of math for the blind, which actually complicate the issue. In the US, work such as the MathGenie project [Karshmer - 2005] has been aimed at providing support for higher level math, but is not easily used in teaching the basics. The AutOMathic block system is designed to fill this void. According to the literature [Ingersoll - 2003], the United States is experiencing a paucity of welltrained math and science teachers in the K-12 system. The problem is exacerbated when we consider the coincident shortage of well-trained special education teachers in the Unites States. The bottom line is quite predictable: the probability of a visually impaired student receiving a quality math education is low.

24

First International Conference on Technology-based Learning with Disability In the remainder of this paper, the AutOMatic block system is described in terms of its use, its potential benefits and the research issues related to its design and development. While reading these descriptions, please keep in mind that the AutOMatic system is designed to: 1. Be a tool used throughout the K-12 system from the earliest days of arithmetic through algebra. After this point, there are other assistive technologies that will provide support for the visually impaired math student. 2. Allows the student to practice solving arithmetic and math problems regardless of the presence of qualified special education teachers, math teachers or qualified parents. 3. Allow the student to learn through practice at virtually any time and in any place. 4. Be useful in virtually any country in the world. 5. Be built of off-the-shelf technology, which will keep its cost low.

A Brief Overview of the AutOMathic Blocks System The concept of the simple Braille math block is extended and enhanced through the use of computers. Not super computers, but rather everyday home computers. The idea is simple: automate the process of tutoring using Braille based math blocks. Using either a simple physical interface, or a virtual haptic interface, the student can become an independent learner with an automated teacher. The Physical Version A simple touch sensitive device is used to present the math problem and then to track the student’s progress in solving the problem. The teacher, parent or student, simply lays out the math blocks on a scored grid (scored to insure alignment of the blocks). Once the problem is laid out, the student has access to more blocks to place on the grid to represent the answer to the problem. Each block contains a bar coded label which when selected, and placed in a scored location on the touch tablet, informs the computer about which number, operator or letter was placed in which position. During the setup and solution phases of the learning process, the computer builds and maintains an internal model of the problem’s current status. The students can add new blocks from his/her block reservoir or remove blocks to be returned to the reservoir. In all cases, the status of the internal representation is updated and advice can be tendered. Movement on the scored tablet is tracked by its touch sensitive nature while movements of blocks in or out of the reservoir are tracked by the special identity tags attached to the blocks. The student, will of course use the Braille tags on the blocks and get verbal feedback at appropriate times. In order to remove or replace a block in the reservoir, the block will pass in close proximity to a small sensing device (barcode reader). In this manner, the current state of the scored tablet and the reservoir can be tracked.

25

First International Conference on Technology-based Learning with Disability The touch tablet will contain Braille-based column and row identification tags, which will be used as a coordinate system conceptually shared by the student and the system much as we do in spreadsheet applications. A flow diagram of the physical block placement process is shown in the figure above and the detailed elements of that figure shown below.

Figure 3: Arranging Blocks in the Physical Version of the System

Figure 4: Arranging Blocks in the Virtual Version of the System

As the blocks are being placed, the computer gives verbal response to the student. For example, in presenting the answer to a simple arithmetic problem, the student forgets to carry from the previous column. The computer would verbally explain that an error has occurred and tell the student generally what went wrong (see Figure 2 above). The Virtual/Haptic Version As this version of the system exists solely in virtual space, the basic procedures of setup and block movement are somewhat different. Because there are no physical blocks, the student or teacher needs only to enter block identifiers via the keyboard to mimic the action of removing one from the reservoir. Once a block is selected, the user is able to place it on the virtual scored tablet by entering the coordinate of the block in virtual space. All errors in this phase will mirror the systems responses found in the physical version.

26

First International Conference on Technology-based Learning with Disability Once the problem is built in the virtual space, the student uses a force feedback device2 to study the block layout and make block movements, replacements or removals under haptic control much as in the physical version of the system. All blocks will be displayed on the computer screen to help the sighted tutor, teacher or parent. Tutoring and Solution Phases When the problem layout phase is complete, the student, teacher or parent will inform the system to start the problem solution phase of the process. At this point several internal actions will take place, including the following (see Figure 5 below) 1. The system will enter the Internal Representation Module to check that the problem is in proper form and construct a canonical representation of the block layout. The blocks are initially stored as a matrix which matches the block layout previously entered. a. The system will read the problem to the user allowing the student to i. Accept the problem and start on the solution ii. Put the student back into the problem build phase to change the basic problem b. The matrix representation is converted to MathML or other canonical representation Control is now passed to the Tutoring module where the student and the system work together to find the correct solution to the exercise previously entered and converted to its canonical representation. This is the key phase of the process and requires the bulk of the team’s effort to build. This effort is described in detail below. For the moment, a simple description of this module is appropriate (see Figure 6 below). At this point in the process, the problem has been laid out and accepted as the exercise specified for solution. Now, the system is waiting for action by the user to drive its next actions. It is now a strictly event driven operation. Whether through physical or virtual change to the AutOMathic blocks, this module carries out the following activities (see Figure 6 below). 1. The changed block layout is a. Converted to the canonical representation b. The canonical representation is sent to the equation solver3 c. Control is passed to the User Feedback element which supplies the tutoring help d. The system then waits for another change 2. If the data returned from the equation solver indicate a correct solution, the student is informed and the session terminates.

2

The project will test a variety of such devices, particularly the new low-cost haptic mice now on the market. 3 There are numerous commercial and public domain equation solving programs available for use in this application. The exact form of the canonical representation will be a function of the equation solver employed.

27


Figure 5. Internal Representation Module

Figure 6. Tutoring Module

Design and Development Challenges In its operational form, the AutoMathic systems seems quite straight forward. But, “under the hood,” there are a number of difficult challenges. In this section of the paper, an overview of some of the most interesting of these challenges are discussed. The project will be designed, researched and developed by the three partner institutions which will each take the lead on a singe aspect of the research.

28

First International Conference on Technology-based Learning with Disability Current Project Status At the moment, a working prototype of the physical version of the project has been implemented. Currently supported functions include addition, subtraction, simple multiplication and simple division. Input to the system is accomplished through the use of small blocks containing Braille notation, bar code labels and visible characters. Using these blocks, a barcode scanner and a touch sensitive pad, the student is able to do simple arithmetic problems with a limited amount of tutoring help. See Figure 7.

Figure 7. Barcode Scanner and Braille block

The cost of these items is quite low, and they offer very accurate and robust performance. The cost, for example, of a set of Braille blocks and a scanner is under $50. We have recently purchased a touch pad (less than $500) which is currently fully functional in the system. During the coming summer, the first author of this paper will spend a month in Finland working on the development of the virtual system with the team at the University of Tampere. The internal data structures, algorithms, integration of the equation solving software and tutoring modules will be developed under the lead of the team at the University of South Florida in cooperation with the California and Finish partners. References

Edwards, D. N. , McCartney, H., Fogarolo F., “Lambda: A multimodal approach to making mathematics accessible to blind students,” ACM SIGACCESS Conference on Assistive Technologies, Portland, Oregon, pp. 48-54, October 2006 Gill, J., Information Resources for People Working in the Field of Visual Disabilities, Royal National Institute for the Blind, 2006 Ingersoll , R., Is There Really a Teacher Shortage?, Center for the Study of Teaching and Policy, University of Washington, September 2003 Karshmer, A., Bledsoe, C., and Stanley, P., “The Architecture of a Comprehensive Equation Browser for the Print Impaired, Proceedings of the IEEE Symposium on Visual Languages and Human Centric Computing, Dallas Texas, September, 2005 Karshmer, A., Gupta, G., Annamalai, N., et al., UMA: A System for Universal Mathematics Accessibility, Proceedings of the ASSETS Conference, 2004 Raisamo, Roope, Hippula, Arto, Patomäki, Saija , Tuominen, Eva, Pasto, Virpi and Hasu, Matias; Testing usability of multimodal applications with visually impaired children. IEEE Multimedia, 13 (3), IEEE Computer Society, 2006, 70-76.

29


AccessScope Project: Accessible Light Microscopy for Students with Disabilities Bradley S. Duerstock Center for Paralysis Research Purdue University Wamiq M. Ahmed School of Electrical and Computer Engineering Purdue University John Cirillo Center for Paralysis Research Purdue University J. Paul Robinson Dept. of Basic Medical Sciences Purdue University ABSTRACT An integrated accessible microscopy workstation, called AccessScope, was developed to allow persons with upper limb mobility impairments and low vision to perform light microscopy without requiring assistance. The ability to operate a light microscope independently is crucial for students to acquire activity-based learning experiences in science curricula. Additionally, independent microscope operation is important for scientists with disabilities to be able to conduct research and pursue a career in the physical and life sciences. The AccessScope workstation is flexible enough for different laboratory-based courses in the sciences and sophisticated enough to perform graduate-level research. AccessScope allows students and scientists with mobility and visual impairments to perform all functions of a brightfield and fluorescent light microscope from loading slides onto the stage to the acquisition of images of slides. INTRODUCTION The light microscope is one of the most common laboratory tools employed by a science student or scientist. Unfortunately, many students with disabilities cannot use a conventional light microscope without requiring assistance from classmates or lab assistants. Being able to control a microscope by oneself provides students with an active learning experience that is necessary

30

First International Conference on Technology-based Learning with Disability for better understanding microscopical imaging concepts and science subjects, like histology, cell biology, botany, and materials science. Performing hands-on experimental techniques and actively investigating scientific concepts is critical to knowledge acquisition and retention (National Research Council, 1996). Thus, activity-based learning is a standard of teaching science at all educational levels (Weld, 1990; Norman et al., 1998; Salend, 1998; Mastropieri et al., 1999; McCann, 2000; Miner et al., 2001). Being able to independently manipulate laboratory equipment allows students with disabilities to pursue educational studies in the science, engineering, and technology (SET) fields and eventually to seek a career in the sciences (Alston and Hampton, 2000; AAAS, 2002). The proportion of undergraduate students with disabilities enrolled in the SET fields is moderately less than able-bodied students (NCES, 1997). However, less than a one-third of undergraduate disabled students compared to their nondisabled college classmates continue to study the life and physical sciences in graduate or professional schools, such as medicine, optometry, and pharmacy. Likewise, less than half of undergraduate students with disabilities studied engineering and computer science in graduate schools compared to able-bodied undergraduate students (NCES, 1997). Thus, during college very few disabled undergraduate students pursue the SET fields on to graduate and professional schools and eventually as a career (NSF, 2001). Previous attempts to make light microscopes accessible have solved only one or a few technical problems, such as viewing microscope slides on a computer monitor or projection screen instead of through the eyepieces or reengineering the focus knobs to be more accommodating (Scruggs and Mastropieri, 1994; Hawke, 2004). We propose an integrated approach to control all functions of a light microscope through a single computer interface, including viewing specimens, focusing, magnification, illumination, condenser control, stage movement, and loading slides. RESEARCH FINDINGS The AccessScope workstation employed a personal computer (PC) to control the hardware components of the AccessScope workstation (Fig. 1). The PC could be adapted with different assistive technology (AT) software and peripheral input devices, such as mouse, trackball, voice recognition, and/or keyboard, to accommodate individuals’ preferences and disabilities (Lau and O’Leary, 1993). The PC ran Windows® operating system to facilitate the use of third-party commercial AT software. Commercial hardware components were used as much as possible to decrease costs and aid in reproducibility of this workstation by others. Figure 1. Components of the AccessScope Workstation. A. A personal computer served as the user interface to control the components of the AccessScope workstation. Users controlled the PC with different input devices, like a keyboard, mouse, or trackball as shown. B. A motorized Olympus BX61™ research microscope was used for automation of microscopy features, such as changing objectives and condensor. C. Specimens are displayed on the computer monitor using a microscope digital camera, eliminating the need to view through the microscope eyepieces. Images of specimens can be stored and managed on the computer for further anatomical study. D. A motorized stage was

31

First International Conference on Technology-based Learning with Disability used to position the slide in X and Y directions. E. The microscope could also be partly controlled through a separate buttonpad. F. An enlarged view of the bulk slide autoloader, which mechanically loads slides stored in two cassettes holding up to fifty slides. G. A separate joystick commanded the movement of the stage. Stage movement can also be controlled through our AccessScope control software. H. Enlarged view of the manual focus knob that was constructed to provide an additional means of fine focusing besides using the software. A large three-spoked knob that does not require finger dexterity was positioned in front of the microscope within users’ reach replaced the original fine focus knob. I. The AccessScope workstation was placed on an automatic table that raises and lowers to the users' preference. Manual and electric wheelchairs and chairs can significantly vary in height.

Accommodation of Users with Mobility Impairments AccessScope was initially designed to be independently operated by individuals with mobility impairments lacking fine or gross upper limb movements. Participants with various upper limb mobility impairments, including those from head and spinal cord injuries, parkinsonism, poliomyelitis, rheumatoid arthritis, and amputation, were able to successfully use AccessScope to perform common microscopy activities without assistance. Participants were able to operate 32

First International Conference on Technology-based Learning with Disability the AccessScope PC using a variety of input devices, including mouse, trackball, and/or keyboard. Windows® provided AT applications, such as StickyKeys™ and MouseKeys™, which further assisted users. StickyKeys™ was useful, not only, for typing also for using hotkeys that required simultaneously pressing a modifier key plus a standard or function key. Participants that lacked fine motor control or had tremors often relied on MouseKeys™ to click on small buttons and drag slider bars, since MouseKeys™ permits incremental movements of the pointer along with clicking through the keypad (Duerstock, 2006). Accommodation of Users with Low Vision The evaluation of AccessScope has recently begun with participants with visual impairments or low vision, including those with retinitis pigmentosa, macular degeneration, and nystagmus. AccessScope has been expanded to assist users with low vision because of their ability to proficiently use Windows® PCs with AT software and hardware and locate general histological structures in a view window on a computer monitor. We believe that these are the two requirements that users with visual impairments must meet to successfully use AccessScope. Preliminary studies have been positive for controlling AccessScope using JAWS® screen reader and ZoomText® screen magnifier software applications. The use of text labels for individual and groups of software buttons in the graphical user interface (GUI) made identification possible for low vision users via JAWS®. Keystrokes were used to execute and navigate through the GUI controls (data not published). Design Features for AccessScope Software A commercial application was previously used to control AccessScope. This application did not operate every feature of the accessible microscopy workstation, including stage movement, which had to be controlled with a separate physical joystick (Fig. 1E). Also, the GUI was not designed to accommodate users with mobility or visual impairments. Though, programmed ‘hotkeys’ could be employed to automatically apply the correct condenser, illumination, and filter wheel when an objective was changed. However, when individual microscope functions needed to be changed the commercial GUI was not very accessible. Features on an interactive map of the microscope had to be clicked on or off with a pointing device. Focusing had to be accomplished using a slider bar or by directly entering distance values. Currently, we are developing our own AccessScope control software to better meet the accessibility requirements of our target groups. From the participant testing we have accomplished thus far, we were able to identify universal design features for the AccessScope control software that would increase usability for both users with upper limb mobility and visual impairments. Though there are differences in the types of AT software and equipment that users with upper limb mobility impairments and those with low vision prefer to use to operate a PC, as mentioned above, there are similarities in software design features that these two groups share. Though some users with upper limb mobility impairments are proficient using a mouse or trackball, keyboard usage is very high among these individuals for pressing hotkeys and precise

33

First International Conference on Technology-based Learning with Disability maneuvering of the pointer (see Accommodation of Users with Mobility Impairments) (Duerstock, 2006). Users with visual impairments also greatly preferred using the keyboard. Therefore, we will make all the controls of the AccessScope control software operable through keystrokes. Most software applications are navigable by pressing ‘tab’, ‘space’, and arrow keys. Depending on the order of controls in the GUI determines what functions are accessible first. An alternative is to use ‘smart’ keys in the AccessScope application. These are dedicated keys that perform specific functions when in the AccessScope program. For example, the arrow keys are specified to control the left, right, forward, and backward movements of the stage and the ‘page up’ and ‘page down’ keys for fine focusing. Viewing slide specimens through the computer monitor instead of through the eyepieces (Fig. 1G) has overcome a major physical obstacle for both users with motor and visual disabilities. However, knowing where the AccessScope preview window is displaying in reference to the entire slide is difficult. Typically when a slide, which may contain several tissue sections, is placed on the microscope stage able-bodied microscopists determine where they wish to examine by alternately viewing through the eyepieces at low magnification and looking at the slide with the naked eye. This usually requires visually scanning much of the whole slide. Once the area of interest is located then a higher magnification objective is selected and the specimen is refocused. This technique would be impossible for users with mobility and visual impairments to perform. One solution to this problem would be to initially scan the entire slide and create a global image map of the entire slide. This global view would allow microscopists to solely rely on the software for navigating specific regions of interest. Another design feature that we wish to incorporate into AccessScope is automatic and predictive user operation. Automatic functions, like autofocus and autoexposure, decrease the need to manually change some microscope settings. Other settings, such as aperture and light intensity, can also be predefined or assigned to specific conditions, like selecting different objectives and choosing brightfield, darkfield, or fluorescence microscopic viewing. Intelligent user interfaces have been developed to predict users' actions when controlling a software application. By anticipating user’s actions usability could be greatly improved. High predictive accuracies in GUI applications have been achieved (Gorniak and Poole, 2000). During light microscopic evaluation certain procedures can be anticipated to occur sequentially. For example, once fluorescence viewing is selected, the program would assume that the user would next want to adjust the exposure rate. By automatically selecting the exposure slider bar, the user would simply need to adjust the slider bar rather than finding, selecting and then making adjustments saving time and effort. CONCLUSIONS Science requires substantial "hands-on" laboratory practice. Developing accessible laboratory equipment is necessary for students and scientists with disabilities to be able to actively participate in science courses and to conduct independent research. The aim of AccessScope is to allow persons with upper limb mobility and visual impairments to fully control all features of a research-level light microscope without needing assistance.

34

First International Conference on Technology-based Learning with Disability Passively studying histology atlases and databases of micrographs are useful tools for studying histology and cell biology. However, the ability to manipulate scientific equipment grants one a greater understanding of how a device works, what tasks it can perform, and what discoveries can be achieved from its use. The benefits of AccessScope to persons with disabilities are evident. The target group that we have been testing had not been able to independently operate a conventional light microscope before. Though they may have been required to use light microscopes in previous science courses, they had to rely on classmates or paid assistants to help them in their use. In a small survey, subjects with mobility impairments had all intentionally avoided laboratory-intensive classes because of their disability. Additionally, the prospect of being required to use job-specific equipment that would likely be difficult for them to manipulate had been a determining factor whether to pursue a particular course of study or occupation (Duerstock, 2006). By allowing persons with disabilities access to light microscopy, we hope to promote greater inclusion of disabled students in science courses and ultimately in SET careers. Development of AccessScope is ongoing. Future features that we wish to incorporate include operating AccessScope through the internet. Students with disabilities would be able to remotely control AccessScope from their computer from practically anywhere. Remote control promotes distance learning and equipment sharing between several users to reduce hardware costs. REFERENCES Alston, R.J. & Hampton, J.L. (2000). Science and Engineering as viable career choices for students with disabilities. RCB, 43(3), 158-164. American Association for the Advancement of Science (2002). New Career Paths for Students with Disabilities: Opportunities in Science, Technology, Engineering, and Mathematics. American Association for the Advancement of Science, Washington, DC. Duerstock, B.S. (2006). Accessible microscopy workstation for students and scientists with mobility impairments. Asst Technol 18: 34-45. Gorniak, P. & Poole, D. (2000). Predicting Future User Actions by Observing Unmodified Applications in Seventeenth National Conference on Artificial Intelligence (AAAI-2000). Hawke, C.S. (2004). Accommodating Students with Disabilities. New Directions for Community Colleges. Wiley Periodicals, no. 125. Lau, C. & O’Leary, S. (1993). Comparison of computer interface devices for persons with severe physical disabilities. Am J Occup Ther, 47(11), 1022-1030. Mastropieri, M.A., Scruggs, T.E., & Magnusen, M. (1999). Activities-oriented science instruction for students with disabilities. Learning Disability Quarterly, 22, 240-249. McCann, W. S. (2000). Science Classrooms for Students with Special Needs. Teaching Strategies. Journal of Early Education and Family Review, 7(4), 23-26.

35

First International Conference on Technology-based Learning with Disability Miner, DL, Nieman, R., Swanson, A.B., & Woods, M. Eds. (2001). Teaching Chemistry to Students with Disabilities: A Manual for High Schools, Colleges, and Graduate Programs, 4th Edition. American Chemical Society Committee on Chemists with Disabilities. The American Chemical Society. National Research Council. (1996). National Science Education Standards. Washington, D.C.: National Academy Press. National Science Foundation. (2001). Division of Science Resources Statistics, Survey of Doctorate Recipients. Ref Type: Electronic Citation Norman, K., Caseau, D., & Stefanich, G.P. (1998). Teaching students with disabilities in inclusive science classrooms: survey results. Sci. Ed., 82, 127-146. Salend, S.J. (1998). Using an activities-based approach to teach science to students with disabilities. Intervention in school and clinic, 34(2), 67-72. Scruggs, T. E. & Mastropieri, M. A. (1994). Refocusing microscope activities for special students. Science Scope, 17(6), 74-78. U.S. Department of Education. (1997). National Center for Education Statistics. The 1995-96 National Postsecondary Student Aid Study. Weld, J.D. (1990) Making science accessible: Special students, special needs. The Science Teacher, 57(8), 34-38.

36


Using Cosmo’s Learning System (CLS) with Children with Autism Corinna Lathan, Ph.D., Katharina Boser, Ph.D., Charlotte Safos, Cathy Frentz, and Kaitlin Powers AnthroTronix, Inc. Silver Spring, MD; www.atinc.com ABSTRACT A 10-week study using a computerized learning tool called Cosmo’s Learning System (CLS) was conducted with 6 children with autism. Each subject received two half hour sessions a week focused on developing identification of color, shape, number, directions, prepositions, and written words as well as sequencing skills and discrimination of relative size and amount. An assessment was created to allow stringent evaluation of the child’s initial skills and progress in the therapy. All subjects demonstrated improvements on the post-test in all content areas except color identification. Children learning prepositions and direction improved an average of 20%. Similar gains were made in written word reading. Children learning relative size/amount improved but had difficulty understanding the concept ‘same’. Subjects showed an increase in the number of trials completed and required less prompting to respond in the last 4 sessions. Although both verbal and non-verbal children made gains, a greater degree of initial comprehension led to greater success. These first results using Cosmo’s learning system are very promising. Further research with more subjects is proposed. INTRODUCTION Research comparing computer-based with non-computer based instruction for children with Autism Spectrum Disorder (ASD) has demonstrated greater effectiveness when children use the computer than with traditional teacher-only methods (Bernard-Opitz et al, 1990). Some existing research has found that autistic individuals consider interactions with a computer less stressful and more engaging than interactions with people, leading to increased learning (Moore & Calvert, 2000). Whether it is the presentation of material in the context of a computer display, or the decreased social interaction required, children with ASD demonstrate greater accuracy and likelihood to respond using a computer to improve skills when compared with more traditional methods (Chen & Bernard-Opitz, 1993). In short, “computers are successful teaching instruments for children with autism due to their multisensory information, controlled and structured environments …and their ability to individualize to each student” (Hetzroni & Tannous, 2004, pg. 96) Other language-impaired populations (e.g., patients with aphasia) have demonstrated a number of impressive language-based improvements as well as maintenance of gains using computerized

37

First International Conference on Technology-based Learning with Disability instruction (Linebarger et al.,2001; Boser et al., 2000 ). However, only a few recent studies have investigated the effectiveness of computers for improving language and communication skills in ASD (Yamamoto & Miya, 1999; Hetzroni & Tannous, 2004). Improving communication and language in autism may require a combination of different techniques, although careful analysis and comparison of available therapies indicates that some form of discrete trial training or Applied Behavior Analysis (ABA) format forms the best basis for children with ASD who are learning new skills (Heflin & Simpson, 1998). The CLS can easily be adapted to an ABA format. The hallmarks of ABA: measurement of behaviors (i.e., responses made), antecedents to behaviors (i.e., stimuli presented) as well as ongoing performance can all be monitored in CLS. While the computer delivers immediate responsebased feedback, the therapist can also adjust and/or amend this feedback as necessary (e.g., delivering a stronger food-based reward). One previous study used an extensive computer-based assessment tool to evaluate ABA-style vocabulary training in a low-functioning nonverbal subject (Boser et al., 2002). This study demonstrated the value of the computer’s storage of all characteristics of a trial including location, features, and names of all responses and items displayed. For the current study, we designed a specific assessment tool to allow access to this valuable information. Although there are numerous advantages of computer-based ABA delivery, including better trial randomization and less errors in performance calculations, studies of computer-based ABA therapy are rare (Gordon, Glatzer, Boser, 2001). Our main objective in this study was to evaluate the efficacy of CLS as a therapy tool for improving receptive language in children with autism. We hypothesized that the preference of children with ASD for computer-assisted instruction and the adaptability of the program to a discrete trial ABA format may predispose them to attend more to the Cosmo’s Play and Learn activities. METHOD GENERAL PROCEDURES All subjects were pre-tested for a week, using both computerized assessments as well as noncomputerized (traditional) evaluations of the skills targeted in the therapy. Traditional evaluation included using manipulative items and table-top tests in a one-to-one format between teacher and student. Since there were no obvious differences in performance between computer vs. non-computerized presentation we used the computer testing to determine the course of therapy and main focus of content. We then completed two half hour therapy sessions a week with the computer, a therapist, and a member of AnthroTronix research staff for a period of 8 weeks. A final week was dedicated to post-therapy assessments, for a total time at the school of 10 weeks. The research staff maintained the computer settings, collected additional data using data sheets, video-taped the sessions for later analysis, and downloaded data for backup and integration into a central database. Video tape recordings were catalogued and copied as a further back-up of each session. SUBJECTS A group of 7 male children with autism, 6-13 years old, were targeted to participate. Based on prior neuropsychological testing, all subjects had a diagnosis of Autism/PDD (made through DSM IV criteria) and a cognitive age approximately 2 -5 years behind their chronological age.

38

First International Conference on Technology-based Learning with Disability All currently attending the Linwood School in Ellicott City, Maryland. Five of these children were 10 and under and two were 12-13 years of age. Three had significant language output and two were starting to learn to recognize some written words. However, the other four were struggling with productive and receptive vocabulary. Permission to participate was granted by the parents of the children through their signature of an IRB-approved letter. Privacy was maintained by creating subject IDs and storing the corresponding names and background data in a separate location.

Subject ID AC-01* AD-02

AGE (years;mos) 6;7 12;10

Table 1: Subject Table LANGUAGE Comments Verbal Single words

AI-03

11;05

Non-verbal

AJ-04

13;10

Echolalic,

AM-05°

9;04

Non-verbal

AV-06*

10;08

Verbal

AZ-07

10:01

Prompted words

Attention problems Whisper-voice, some comprehension Very active; poor attention; difficult to test/evaluate Lethargic; little comprehension Dropped out; poor testing High functioning, working memory problems Attention problems; difficult to test due to dietary illness

CLS Activity Completed(p. 5) All 1, 2, 3, 4, 6, 7 1, 6, 7

1, 6, 7 0 All

1, 6, 7

*indicates that the subject completed all difficulty levels of the tasks ° subject dropped out of study

COSMO’S PLAY AND LEARN SOFTWARE Cosmo’s Play and Learn software is a family of educational computer packages that target children of developmental ages 2-8. CLS consists of a computer interface device (a monitor and mission control) and educational software. Using Mission Control, a keyboard-sized interface with built-in microphone and four aFFx activators designed to interact with a computer or control CosmoBot, children empower a virtual robot to explore a playground filled with activities. The main character of the software is Cosmo, a virtual robot, who motivates and engages the children. Created in conjunction with educators, therapists, and assistive technology practitioners, the software allows the facilitator to target specific goals and track the child’s progress. CLS has a comprehensive curriculum guide and magnetic manipulatives to reinforce content areas present in the software. The CLS software has data collection that tracks general therapeutic and educational activities. COSMO’S ASSESSMENT: Cosmo’s Assessment was developed by Dr. Katharina Boser to provide a precise, in-depth measure of student progress before and after using Cosmo’s Play and Learn System (CLS). The assessment tool was created to measure knowledge of all content areas of the computer-based therapy. The tool is innovative because it takes into account fundamental issues relevant to

39

First International Conference on Technology-based Learning with Disability working with children with severe learning impairments, it can be modified for other content goals, and provides accurate measures for reporting progress. One learning principal that the assessment addresses is the problem of response interference that lower functioning subjects exhibit in learning new skills (for detailed discussion of interference in stimulus control see Green, 2001). For example, a subject may continue to choose a learned item and ignore new target stimuli. Alternatively, they may choose items at a location that was previously reinforced. One method to counteract such interference, is to avoid consecutive testing of the same item. In this study, we adopted best practices for establishing conditional discrimination in teaching procedures in this population as described in Green (2001) among others. Cosmo’s Assessment is computer-based and facilitated by an adult. The assessment measures:1) identification (ID) of color, 2) color sequencing 3) shape ID, 4) number ID, 5)identification of the 28 written text words corresponding to the content, 6) discrimination of relative size (bigger/ smaller/same) and 7) relative amount, e.g., more/less/same, 8) directions (up down right left); 9) identification of prepositions (inside, outside, behind, in front, over under). Figure 1 below provides a visual depiction of a typical screen for the 9 assessments. Target items were randomly presented in four different locations, preventing chance correct responses with a small sample. The output data provide information regarding each trial, including not only the target location and content but also all other response options. This allows the tester to prevent guessing and to evaluate response strategies. For children with lower ability level a 2-item version of the task was shown in which 2 of the 4 boxes displayed were empty.

Color ID /Sequence

Relative Amount

Shape Identification

Number Identification

Relative Size

Directions Figure 1: Assessment Screens

Written word/ Text

Prepositions

Cosmo’s Assessments consisted of 10-12 trials unless the child made three consecutive errors. If three consecutive errors are made, the facilitator exits the program. Some children received a choice of rewards to maintain interest in Cosmo’s Assessment and the computer as necessary. These rewards were initially presented every trial and then faded as the child became increasingly engaged. Breaks were provided as necessary. No feedback was provided for responses except to maintain attention to the task (e.g., “good sitting”, “good looking”).

40


COSMO’S PLAY AND LEARN (CLS) THERAPY Therapy consisted of 2 half hour sessions with the CLS for 8 weeks. Therapists at Linwood worked with our staff to use the system and help the children with any possible behavior issues. Linwood therapists transported children to and from their individual classrooms to a separate quiet area where the computer system was maintained. The children were encouraged to use the “free play” sessions as breaks where needed. During the therapy with CLS, text remained ‘on’ and available during all training sessions (Figure 2). Children were allowed maximum time to respond; but with adjustment of timing of Cosmo’s prompts for each child/situation. Therapists were encouraged to provide cues as they would during a normal therapy session to keep the child engaged and on task. Teachers from the Linwood School were allowed to remain in the room, aiding only when a child’s attention waned significantly. Over time, Cosmo provided the main source for cues and help to the child. The program incorporates in a lot of reinforcement for correct responses (e.g., sound, movement of animated characters or symbols). Feedback in CLS was targeted, immediate and leveled. It increased in amount and informativeness as the child continued to make an error (less to more prompting). First, a ‘try again’ response was given, then the question is repeated and a visual cue is given. After a third incorrect response, Cosmo provides the correct answer and a new trial begins. If the child needs more than 3 ‘give away’ cues (highest cue), the therapist was encouraged to move on to either ‘freeplay’ for that module or a new module, or perhaps try a ‘break’ using manipulatives. Otherwise, 10-12 trials in each module was attempted

1. Blowing Bubbles

2. Bubble-a-Tron

5. Playing Tag

6. Number Jumping 7. Puzzle-a-Tron Figure 2: Therapy Screens

3. Bubble Recipes

4. Popping Bubbles

8. Free Play

The therapy involved a variety of activities in each of the content areas. The following descriptions provide a brief description of the activities for each content area. See Figure 2 for visual depiction (screen shot). Table 1 lists the activities each student completed.

41

First International Conference on Technology-based Learning with Disability 1. Blowing Bubbles: The child learns colors by blowing red, yellow, blue or green bubbles when they press on the correct mission control activator. Difficulty is increased by asking the child respond to two color names at once (color sequencing). 2. The Bubble-a-Tron: In the Blowing Bubbles with the Bubble-a-Tron activity, bubble sizes help illustrate the sometimes subtle differences of “bigger,” “smaller,” and “same.” 3. Bubble Recipes: Following auditory cues regarding relative amount the child must measure the bubble solution very carefully into beakers over the Bubble-a-Tron. 4. Popping Bubbles: Children must hold down the correct mission control activator until Cosmo pops the bubble in the correct location: up, down, left and right. 5. Playing Tag: A virtual game of “Tag” introduces and/or reinforces basic prepositions related to space: under, over, inside, outside, in front of and behind. Children must press the correct activator to move Cosmo in the correct direction to catch his friends, hiding in the targeted location. Task difficulty is increased by changing features and numbers of distracters. 6. Number Jumping: Friends in clusters of 1, 2, 3, 4 or 5 fly above each Space Animal. Cosmo asks the child to help his friends jump over a certain number of flying friends. Once they have been caught, they will do a “counting off” animation and fly off screen. Task difficulty is increased by changing features and numbers of distracters. 7. The Puzzle-a-Tron: Puzzles created by a Puzzle-a-Tron introduce the basic shapes: circle, square, triangle and rectangle. Cosmo asks the child to help his friends jump over the correct shape. Task difficulty is increased by changing features and numbers of distracters. Free Play: Each module provides an opportunity for the child to explore and learn without assessment, driving their own learning (e.g,. create your own bubbles or recipes, etc.). USE OF ‘CAPS’ Caps were designed to cover the colored mission control activator boxes for identification tasks where color was not the specific target, i.e., the shape and number tasks. The caps focused the students’ attention away from the colored buttons and toward the corresponding ‘friends of Cosmo’ lined up on the bottom of the screen in these tasks. The caps were small clear plastic tops into which a small picture of Cosmo’s friends was inserted. By eliminating any distracting information, the child was better able to make a direct link between making the specific friend ‘jump’ to catch the correct number or shape and the activator button they needed to press. RESULTS (please refer to figures 3-6 on the following page) Most subjects demonstrated improvements at post-testing in at least one content area except color identification (Figure 3). Note that unless otherwise indicated the task was a 2-choice discrimination (i.e., chance performance is 50%). Subjects learning relative amount and size (CLS therapy # 2 & 3) improved in understanding relative terms (e.g., bigger, more) but not ‘same.’ One nonverbal subject (AD) demonstrated a large percentage increase in number ID (from 34%-92% correct) and shape ID (59%-83% correct). AC and AV, who participated in preposition and direction therapy (CLS # 4 & 5) improved an average of 20%. Similar gains were made in text reading (Figure 4). AV demonstrated an equally large percentage increase in his ability to repeat color sequences at post testing from 35%-83% correct. Subjects AV and AC who completed relative amount (CLS #3) struggled with the notion ‘same’ but improved for ‘more’ and ‘less’ (e.g., from 40%-100% correct). Unlike AD (Figure 5), both AC and AV remained very good at understanding relative size (CLS #2) terms ‘bigger’ and ‘smaller’ (92100%), but both struggled with ‘same’ (50-67% at post-test).

42

First International Conference on Technology-based Learning with Disability Subjects AD, AJ and AI demonstrated an increased number of trials that required no prompt or one prompt (e.g., “try again”) in the final 4 sessions of therapy. Figure 6 showed the difference between the number of trials requiring prompts for the first 4 sessions (‘pre’) and those in the final 4 sessions (‘post’). The amount of trials completed per session increased from pre to post sessions for most subjects. AJ’s trials increased from 19 to 40 for number ID. AI’s trials increased from 65 to 127, AJ’s from 74-115 trials and AD’s increased from 44 to 128 trials for Shape ID.

43


Shape ID Pre/Post Changes

pre post

AC-01

AD-02

AI-03

pre post

AD-02

AI-03

AJ-04

pre post

AC-01

AZ-07

AD-02

AI-03

AJ-04

AZ-07

Subject ID

Color Assessment

Shape Assessment

Figure 3: Identification tasks; number, shape and color

Directional Awareness

Text/Reading Ability

Prepositions

*4 choice task 100%

100% 80% 60%

pre

40%

post

20%

60%

pre

40%

post

20%

Perce n t C o rrec t

80% P ercen t C o rrect

Perc en t C o rrec t

*

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

Subject ID

* 4 choice task

Number Assessment *4-choice task

ColorID Pre/Post Changes

*

AC-01

AZ-07

Subject ID

* 4 choice task

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

Percent Correct

Number ID Pre/Post Changes *

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

Percen t C o rrect

Percent Co rrect

RESULTS FIGURES

0%

0%

AC-01

AD-02

AC-01

AV-06

Subject ID

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

pre post

AC-01

Subject ID

AV-06 Subject ID

Directions

Prepositions

Text/Written word Figure 4: Directions, prepositions and text improvement for AC and AV

pre post

AC-01

3 colors AC-01

2 colors AV-06

100%

100%

80%

80%

60% pre

40%

post

20%

pre

60%

post

40% 20% 0%

0%

3 colors

bigger

AV-06

Subject ID

Color Sequencing

Percen t C o rrect

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

2 colors

Relative Amount

Relative Size P e rc e n t C o rre c t

Percen t Co rrect

Color Sequencing *4 choice task

more

smaller

AD-02

less

same

more

less

same

AC-01 AC-01 AC-01 AV-06 AV-06 AV-06

AD-02

Subject ID

Relative Size Relative Amount Figure 5: Color Sequencing, Relative Size and Quantity AJ-04 Number ID

AD-02 Number ID

AI-05 Shape ID

pre post

80% 60%

pre post

40% 20%

percent of trials

100% percent of trials

percent of trials

100% 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

80% pre

60%

post

40% 20%

0% no prompt 1st prompt 2nd prompt 3rd prompt

no prompt 1st prompt 2nd prompt 3rd prompt

prompts

prompt

0% no prompt 1st prompt 2nd prompt 3rd prompt prompt

AI-05 Shape Therapy AD-02 Number Therapy AJ-04 Number Therapy Figure 6: Therapy data for AD, AI and AJ

44

First International Conference on Technology-based Learning with Disability DISCUSSION Improvements were found in all areas targeted by the CLS, except color identification. This is not surprising given known research on the difficulty of teaching and learning color terms which are categorical in nature and subject to categorical perception (Bornstein, 1985; Rosch, 1975). Unlike shape and number, color also refers to a characteristic of an object that can change its properties in subtle ways depending on the object. The difficulty subjects faced with understanding the term ‘same’ compared with learning the terms ‘more/less’ or ‘bigger/smaller’ may also be due to the relative conceptual difficulty of the term. Judging sameness or similarity is like an analogy and thus more complicated than making a direct visual comparison such as ‘more’, ‘less’, ‘bigger’, ‘smaller’ (see Gentner & Markman, 1995). The fact that even lower functioning children AJ, AZ and AI made some improvements in receptive knowledge, and learned to respond more over time, using CLS is encouraging. AJ and AI demonstrated these improvements more through the decreased necessity of prompts than in large pre-post gains. AZ was the one child who missed some therapy due to illness, yet he seemed to make the gains toward the end of the therapy. CONCLUSION ASD is a lifelong neurodevelopmental disorder. Early intervention has been prioritized because of evidence that it can make a significant difference to the individual's development (Butter et al., 2003). While even aggressive and early intervention is unlikely to change the diagnosis (Howlin, 2003), it can affect the individual’s eventual success across all core life domains (Butter et al., 2003). A computerized tool such as Cosmo’s learning system has the potential to make great improvements in a child’s quality of life and in the therapists ability to monitor and reward progress. Results from this initial study encourage us to investigate CLS further as a very promising learning tool that engages students in longer learning intervals. The ultimate goal for the students is to transfer skills learned to real-world situations involving activities of daily living (ADL) and everyday interactions with parents, teachers, and peers. Multimodal control, like other enriched computer environments, will add to the vividness of these learning experiences as well as to their ecological validity, which, in turn, is expected to enhance transfer of training (Trepagnier et al., 2006). We are working toward developing the software as part of a larger system that includes CosmoBot, an interactive robot and extension of ‘Cosmo’ from the software (Brisben et. al., 2006). This will allow us to include joint attention, imitation, and other social and communication skills in the child’s learning repertoire. Our results suggest that continued development of stimulus presentation, prompting methods as well data collection and presentation will be crucial in the continued enhancement of CLS as a valuable learning tool for children with autism and those working who work with them. REFERENCES Bernard-Opitz, V., Ross, K., & Tuttas, M. L. (1990). Computer assisted instruction for autistic children. Annals of the Academy of Medicine, 19, 611–616. Bornstein, M. H. (1985). On the development of color naming in young children. Brain and Language, 26, 72-93.

45

First International Conference on Technology-based Learning with Disability Boser, K., Higgins, S., Fetherston, A., Preissler, M. A., & Gordon, B. (2002). Semantic fields in low-functioning autism. Journal of Autism and Developmental Disorders, 32 (6), 563582. Boser, K., Weinrich, M., & McCall, D. (2000). Maintaining verbal production in aphasia: evidence from tense morphology training. Journal of Neurology and Neurological Repair, 14, 105-118. Brisben, A. J., Samango-Sprouse, C., Safos, C. J, Lathan, C.E. Evaluating CosmoBot’s Efficacy as a Treatment Approach for Children with Autism Spectrum Disorder (ASD). Poster Presentation, International Meeting for Autism Research, Montreal, Canada, June 2003. Butter EM, Wynn J, Mulick JA. Early intervention critical to autism treatment. (2003). Pediatric Annual. 32 (10):677-84. Chen, S. H., & Bernard-Opitz, V. (1993). Comparison of personal and computer-assisted instruction for children with autism. Mental Retardation, 31, 368–376 Colby, K. (1973). The rationale for computer-based treatment of language difficulties in nonspeaking autistic children. Journal of Autism and Childhood Schizophrenia, 3, 254–260 Gentner, D., & Markman, A. B. (1995). Similarity is like analogy: Structural alignment in comparison. In C. Cacciari (Ed.), Similarity in Language, Thought, and Perception: Brepols. Gordon, B., Boser, K., & Glatzer, R. (2001). Integrated testing and training platform for lowfunctioning individuals with autism. Paper presented at the International Meeting for Autism Research, San Diego, CA. Green, G. (2001). Behavior analytic instruction for learners with autism. Focus on Autism and Other Developmental Disabilities, 16(2), 72-85. Howlin, P. (2003). Can early interventions alter the course of autism? Novartis Found Symposium. 251:250-9; discussion 260-5, 281-97. Hetzroni, O. & Tannous, J. (2004). Effects of a computer-based intervention program on the communicative functions of children with autism. Journal of Autism and Developmental Disorders, 34(2). 95-113. Heflin, J., & Simpson, R. L. (1998). Interventions for children and Youth with Autism: Prudent choices in a world of exaggerated claims and empty promises. Part I: Intervention and treatment option review. Focus on Autism and Other Developmental Disabilities, 13(4), 194-211. Linebarger, M. C., Schwartz, M. F., & Kohn, S. E. (2001). Computer-based training of language production: an exploratory study. Neuropsychological Rehabilitation, 11(1), 57-96. Moore, M., & Calvert, S. (2000). Vocabulary acquisition for children with autism: teacher or computer instruction. Journal of Autism and Developmental Disorders, 30(4), 359-362. Rosch, E. (1975). The nature of mental codes for color categories. Journal of Experimental Psychology, 1(4), 303-322. Trepagnier CY, Finkelmeyer A, Sebrechts MM, Stewart W, Woodford J, Coleman M. Simulating social interaction to address deficits of ASD. CyberPsychology and Behavior. 2006 9(2): 213-7. Yamamoto, J., & Miya, T. (1999). Acquisition and transfer of sentence construction in autistic students: analysis by computer-based teaching. Research on Developmental Disabilities, 20(5), 355-377.

46


Acknowledgements The authors wish to thank the Department of Education’s Rehabilitation Engineering Research Center on Telerehabilitation for funding the study. We thank especially the participating students at the Linwood School, the teachers and the director, Karen Spence for helping us to conduct our research. We also wish to acknowledge the assistance of Amy Brisben from AnthroTronix,Inc. in administration of the project.

47


Studies on Display/Icon Complexity and Its Relationship to Performance D. W. Repperger D. L. Aleva G. F. Thomas Air Force Research Laboratory AFRL/HEC, WPAFB, Ohio 45433 S. C. Fullenkamp General Dynamics Inc. Dayton, Ohio 45433 C. A. Phillips Wright State University, Department of Biomedical, Industrial, and Human Factors Engineering, Dayton, Ohio 45435 ABSTRACT A summary of current work relating to the design of complex displays for human decision makers is discussed. The results reported here can be applied to other visual rending methodologies to assist in the redesign of complex visual displays to assist people with physical challenges. The present investigations have focused on military iconic technology and how best to reshape the manner in which such renderings are constructed. Of the four studies presented, the first investigation dealt with the identification of human biases, the second experiment provided a rank-order paradigm of complexity of specific iconic features that are perceived by their users. The third study investigated overload and capacity issues in gathering information from complex displays. The last investigation looked at how animation may be a portal to the means of obviating human biased effects and help manage the attention of the decision maker or possibly the physically challenged individual. INTRODUCTION In developing technology-based learning for people who have both physical and cognitive challenges, designs of displays and other information delivery systems must take into account whatever difficulties that technology presents and to best match this technology to the capabilities and needs of the user. For example, a common complaint among the older population is that computer technology/devices are difficult to employ and understand, so they

48

First International Conference on Technology-based Learning with Disability offer little benefit if such technology is not utilized. In studies on human performance, one key attribute is the appropriate display of information to operators tailored to their specific needs. It is all too common, nowadays, in our modern information age to have too much information or information which may not be relevant to the task at hand. The literature on complexity of iconic objects is quite vast (Chernoff, 1973). It includes classical studies to read moving text on CRT screens (Granaas, et al., 1984), tests with Chinese text (Chen & Chien, 2005), visual search issues (Chan, et al., 2005), visual working memory (Logie & Pearson , 1997), workload effects (Mayes, et al. 2001), face recognition (Moscovitch, et al., 1997), cultural differences (Santos, et al. 2005), and effects of lumination contrast (Shieh & Chen, 2005). Studies have also included primates (Trevarthen, 1968) which have a rich history of investigation. One research project examining how to better match levels of information display and the user’s needs and capabilities has been ongoing at the Air Force Research Laboratory at Wright Patterson Air Force Base in Dayton, Ohio involving icon complexity (Repperger et al. 2006a,b,c, 2007a,b). Using information theoretic means, various levels of complexity of icons are being studied. The original impetus of the work was to examine information-theoretic models of how humans effectively glean information from complex iconic objects. The goal would be to have a better idea on how to design complex iconic objects. This research project is not unlike the development of improved displays for technology-based learning involving individuals who may have both physical and cognitive challenges. In order to address the complexity of icons problem, a literature search was first conducted in the area of visualization of information (Repperger, et al., 2006c). The history of this area is quite extensive, going back several thousand years, when people needed to convey information succinctly, via maps and other visual renderings. The areas researched include the historical development of displays, iconic representations, the potential benefit of visual animation to human performance, and short time memory issues with the recollection of information from displays. It was found that there are conflicting views in the literature about certain types of visualizations. For example, the literature suggests that visual animation may have both benefits and detrimental effects with respect to human performance. Animation may be beneficial in quickly grabbing attention, but may also provide unnecessary distraction which may not benefit the task at hand. Since there are numerous complex iconic representations that can be investigated, the initial goal has been to initially focus on military icons, of which a standard data base is well established. One benefit of this research to the US Air Force is to better understand how the military could develop more advanced displays (iconic representations) to enable a commander make military decisions that are more effective (accurately and quickly) when under time pressures or other high workload conditions. The particular approach to examine this problem is through a number of modeling and analysis procedures. The early work of Hick and Hyman (Hick, 1952 and Hyman, 1953) was extremely relevant since they had established limits of human information processing capability from a perceptual point of view. This paper will report on a number of early experiments to develop information overload models and how this affects human performance. The first study (Repperger, et al., 2006a,b) looked within a signal detection theory framework in order classify the complexity of the iconic dimensions (or features) and their rank order of saliency. The term “saliency” means the ability of a certain iconic dimension (or feature) to grab the attention of the user. It turns out that specific iconic dimensions were significantly more salient than others, indicating human-induced biases that affect how information was gleaned off of a complex display such as a high

49

First International Conference on Technology-based Learning with Disability dimensional military icon (Repperger, 2007b). Once a rank-order of the iconic dimensions in terms of their saliency was established, the complexity of the iconic object could then be investigated with the independent variable being the dimension number. In (Repperger, et al., 2007b) accuracy results showed exponential laws of performance versus the complexity of the iconic object. Also, exponential laws were obtained relating received information to time duration. Additional studies have been ongoing to look at the possible detrimental /beneficial effects of animation of iconic features (Repperger, et al. 2007a) versus their static visual rendering counterparts. For the military application, the disruptive nature of animation may be embraced since it quickly directs the attention of the decision maker to key points on a map display. Thus the management of animation on the display may be viewed as the management of the attention of the decision maker. A fourth study looks at the information-theoretic results of Hick and Hyman and casts the present investigation within the context of the early work exploring perceptual overload and the complexity of the iconic object. The independent variable is iconic complexity, which deals with the topic of this paper on how to better design displays/iconic objects to best match the needs and capabilities of the user in a capacity sense (bits/second). The generalization of these results to the area of technology-based learning for people who have both physical and cognitive challenges has many analogies. More details on each of the experiments are now presented. STUDY 1 – IDENTIFICATION OF HUMAN PERCEPTUAL BIAS Figure 1 shows a rendering of the motivation of the first study. A decision maker is looking at a complex map and gathering meta information from an iconic object. This icon describes some information about a military asset on the map. Figure 2 is a representation of such a complex military icon (MIL2525B) with the ten dimensions in one of the two possible binary states. Figure 3 displays the experimental setup to run all experiments with human subjects. Using a signal detection theory framework (Repperger et al., 2006a,b), Figure 4 shows how the sensitivity and specificity measures were constructed from the truth table. In this first study, the complexity of the icon was Information Source

Decision Maker Perceptual Filter

data) (meta

Dynamics?

The Concept of a Human Perceptual Filter

Fig. 1 – The Perceptual Bias Filter

A 10 Dimensional Icon for MIL2525B

Figure 2 – A Complex Military icon

synthesized by choosing the dimensions or features in an arbitrary manner. Figure 5 shows the results of the first study with the sensitivity and specificity plotted versus the complexity of the object (dimension number).

50


True State of the World h1

H1

h0

n1

n2

n4

n3

Human Responses

H0

PD = PHit =

n

, P =

n

Sensitivity = (n1)/(n1+n 1 4), Specificity = (n3)/(n22+n3) FA

+ n4 Table for nData Signal Detection Theoryn-1 Truth Collection 2 +n 3

Experimental Scenario Used to Run Subjects

Figure 3 – The experimental setup

Figure 4 – A signal detection theory framework

Specificity and Sensitivity

From the data in Figure 5 it was obvious that a confound exists in the data. One would expect that as the complexity of the icon increases (going to the right on the x axis), the sensitivity and specificity should decrease in a monotonic manner. However, it is observed that there exists a sharp decrease of specificity at dimension 6 and a concurrent decrease in the performance. Further increases in dimension number (complexity) showed an increase in specificity. This result confounded the data because humans tend to have a perceptual bias and do not perceive all dimensions of complex icons in an equal manner. This motivated the second study to find the exact order of the perceptual

Sensitivity and Specificity versus Number of Dimensions 1.02 1 0.98 0.96 0.94 sensitivity specificity

0.92 0.9 0.88 0.86 0.84 0.82 0

1

2

3

4

5

6

7

8

9

10

Number of Dimensions Study 1 - Sensitivity and Specificity versus Number of Dimensions for the Icon

Figure 5 – Results of the First Study – With A Confound Due to a Perceptual Bias biases of the operator which would describe the dynamics of the perceptual filter illustrated in Figure 1. The second study will provide the correct perceptual hierarchy.

51

First International Conference on Technology-based Learning with Disability STUDY 2- RANK ORDERING THE HUMAN PERCEPTUAL SYSTEM Figure 6 shows new data from the second experiment when the feature or dimension are average across subjects and time 1 0.9 0.8 0.7

Chance Line

0.6 0.5 0.4 0.3 0.2 0.1

ho sti le

no rth

ea st

su rfa ce

re inf or ce d

gr ee na rro w

pla nn ed

6 nu mb er

tas kf or ce

nu cle ar

0

Figure 6 – Results - Second Study to Develop a Monotonic Paradigm of Perception correctly rank ordered in a manner so that performance of each individual feature is fairly ranked with all other features. This diagram has the requisite monotonic nature that as the dimension number increases, the human performance (accuracy = (total correct hits and rejects)/ all trials) will tend to decrease, accordingly. With the dynamics of the perceptual filter quantified by Figure 6, a preliminary investigation on overload capacity was then initiated.

STUDY 3- INFORMATION THEORETIC INVESTIGATION The third study examined the original premise of this work: i.e. how complex should an iconic rendering become before it loses its efficacy? Using 4 different stimulus presentation times, the accuracy was determined and capacities could be determined. In this experiment, all icons had 10 dimensions or features and an equal number of questions were asked on the state of each feature. Since four presentation times were utilized, the four different capacities could be determined as (Log2(10))/ISI multiplied by the accuracy to reflect the number of bits received by the operator. Note ISI is the interstimulus interval. Figure 7 shows the results both as a function of the complexity (dimension number) and also the presentation time. The results indicate an accuracy rule that decreases with complexity (dimension number) in an exponential manner. The regression coefficient had an R2 value of 0.8681 when averaged over all subjects and features as indicated in Figure 7.

52


Results of Study 3 – Information Theory Investigation averaged across subjects and features

Information Theory - Phase 3

1

0.01

0.98 0.96 0

percent correct

0.94

0

2

4

6

8

10

125 ms 250 ms

0.92 0.9

500 ms

0.88

-0.01

1000 ms

0.86 0.84

0.8 4dim

5dim

6dim

7dim

8dim

9dim

10dim

Log_10 of Accuracy

-0.02

0.82

-0.03

-0.04

Rule 2 – Exponential dependence of accuracy with dimension number:

y 2 = K 3e

− ( α 2 )( D )

y = -0.0067x + 0.0293 R2 = 0.8681

-0.05

-0.06

-0.07

Results of Study 3 – With 4 Presentation Times

Dimension number

Figure 7 - Results of Study 3 for Information-Theoretic Capacity measures The last study addressed the question what value could animation add to help alert the operator to a higher level of vigilance? STUDY 4 – EFFECT OF ANIMATION ON MANIPULATION ATTENDTION From study 1, it was determined that there exists a saliency bias filter which describes the nonlinear dynamics of the human perceptual system to correctly identify features. Figure 6 actually identified the dynamics of this perceptual filter for the MIL 2525B standard within the USAF. The question that is immediately raised is what happens if we animate one of the weakest features and see if we can raise its saliency? The suggestion was that there may be a finite resource model involved, in which there would be a tradeoff effect in the sense of a zero sum game. That is, if we increased saliency with one feature, we may reduce the saliency of another feature. Figure 8 shows how the less salient feature (icon dimension = number 6 or 9) was flashed in two different ways. One animation method flashed the feature with a 100% duty cycle (black and white at about 3 times a second) on a background either red or blue. The second method flashed the feature only 50% of the duty cycle by having the number either black or to disappear completely. Figure 9 shows the results of this last study. Two outcomes are noted: (1) The 100% duty cycle case resulted in better performance than either the static or the 50% duty cycle animation. Also, (2) As the performance improved on the weak feature, there appeared to be a zero sum game effect in the sense that another feature

53


Accuracy ( %correct )

Effects of Feature Animation on Feature Saliency 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

no animation black to blank black to white

9 st ) th air or # mical arrow there uced e (y/n ce or riendly or we r sou r f east he r red eally or red k forc surfa th o c o r r o e o r n til d tas r r o een s e o a o c r d le gr h nuc nne reinfo pla #6

Flash this feature.

Figure 8 – Flashing the feature

Feature

Figure 9 – Results of the animation

suffered significantly though this modification. Thus a finite resources model seems to apply, i.e. if one feature is animated to improve performance, this may result in another feature having a loss of performance. Also in Figure 9, the changes denoted by the red circle show the benefits of animation over the static renderings. The blue circles indicate that the other features (non animated) lost some of their performance. This is pinpointing of a finite resource model for the human. SUMMARY AND CONCLUSIONS From the results of four studies on military iconic technology, it was determined that humans have perceptual biases that need to be identified. Once these filter dynamics are determined, then to fairly construct icons that deliver appropriate amount of information, and complexity, they should be synthesized inversely in the order described in Figure 6. If an iconic object becomes too complex, then it would be appropriate to “drill down” to a less complex rendering. This technique should extrapolate to people who have physical/cognitive challenges since their levels of information capacity may differ from the normal groups considered in these studies. Finally, it is demonstrated that animation of a less salient feature will add value in increasing the accuracy to correctly respond to that feature, but at the same time, there may be a finite resource model in place which tends to state that increasing performance on one feature may reduce the accuracy to correctly perceive another feature on a complex iconic object. REFERENCES Chan, A. H. S., Yu, R. F., & Courtney, A. J. (2005). Quantifying visual field shape for improving accuracy of search performance prediction. Perceptual and Motor Skills, 100, 195-206. Chen, C-H & Chien, Y-H. (2005). Effect of dynamic display and speed of display movement on reading Chinese text presented on a small screen. Perceptual and Motor Skills, 101, 865-873. Chernoff, H. (1973). The use of faces to represent points in k-dimensional space graphically, J. Am. Statistical Association, 68 (342), 361-368. Granaas, M. M., Mckay, T. D., Laham, R. D., Hurt, L. D., & Juola, J. F. (1984). Reading

54

First International Conference on Technology-based Learning with Disability moving text on a CRT screen. Human Factors, 26, 97-104. Hick, W. E. (1952). On the rate of gain of information. Q. J. Exp. Psychol., 4: pp.11 -26. Hyman, R (1953). Stimulus information as a determinant of reaction time. J. Exp. Psychol., 45, 188-196. Logie, R. H., & Pearson, D. G. (1997). The inner eye and the inner scribe of visual working memory: evidence from developmental fractionation. European Journal of Cognitive Psychology, 9, 52-55. Mayes, D. K., Sims, V. K., & Koonce, J. M. (2001). Comprehension and workload differences for VDT and paper-based reading. International Journal of Industrial Ergonomics, 28, 367-378. Moscovitch, M., Winocur, D., & Behrmann, M. (1977). What is special about face recognition? Nineteen experiments on a person with visual object agnosia and dyslexia but normal face recognition. Journal of Cognitive Neuroscience, 9, 555604. Santos, F. H., Mello, C. B., Bueno, O. F. A. & Dellatolas, G. (2005). Cross-cultural differences for three visual memory tasks in Brazilian children. Perceptual and Motor Skills, 101, 421-433. Repperger, D. W., Aleva, D. L., Thomas-Meyers, G. & Fullenkamp, S. C., (2006a). Visual perception saliency investigation with complex icons. AIAA 31st. Annual Dayton-Cincinnati Aerospace Science Symposium, Dayton, Ohio. Repperger, D. W., Aleva, D. L. Thomas-Meyers, G. F. & Fullenkamp, S. C. (2006b). Studies in icon complexity and visual displays. The Ohio Journal of Science, 106 (1), A-54. Repperger, D. W., Kuperman, G., Thomas-Meyers, G., Aleva, D. L., & Fullenkamp, S. (2006c). A compendium on glyph/icon research including MIL 2525B, AFRL-HEWP-TR-2006-0149. Repperger, D. W., Phillips, C. A., Thomas-Meyers, G., Aleva, D. L., & Fullenkamp, S. C. (2007a). Benefits of attention disruption with animated iconic objects. The Ohio Journal of Science, 107(1), A-28 Repperger, D. W., Thomas-Meyers, G., Aleva, D. L., Fullenkamp, S. C. & Phillips, C. A. (2007b). Investigating display icon complexity via information-theoretic constructs, The 78th Aerospace Medical Association Meeting, New Orleans, LA. Shieh, K-K & Chen, M-H. (2005). Effects of display medium and luminance contrast on concept formation and EEG response. Perceptual and Motor Skills, 101, 943-954. Trevarthen, C. B. (1968). Two mechanisms of vision in primates. Psychologische Forschung, 31, 299-337.

55


Combining Gaze Interaction and Facial EMG Clicking Andrew Michael Junker Brain Actuated Technologies, Inc. 1350 President Street, Yellow Springs, Ohio 45387 USA [email protected] John Paulin Hansen IT University of Copenhagen Rued Langgaards Vej 7 2300 Copenhagen S Denmark [email protected] ABSTRACT For a user to control an eye gaze system they must be able to control their eyes and keep their eyes open. In most cases when this is possible the user will have some control of their facial muscle activity as well. Facial muscle activity, the electomyographic (EMG) signal, may be used for clicking to enhance a gaze communication system (Surraka et. al., 2004). The EMG signal can provide a fast click that would overcome problems associated with dwell or eye blink click. The EMG click may serve as a back up for gaze control when the portability of a user’s eye gaze system is a problem or if the user should loose the ability to control their eyes. GAZE INTERACTION Gaze interaction systems are powerful communication tools for people who are not able to control their hands. The technology has been available for more than 20 years (Majaranta and Raiha, 2000) and is still improving. It supports severely disabled people such as Amyotrophic Lateral Sclerosis (ALS) patients who have lost their ability to control their hands. They can communicate with their family by gazing at a particular part of a computer screen with characters or commands. Gaze communication systems consist of an eye tracker (a camera pointing towards the users eye) and an interface. On-screen keyboards (e.g., “Point for Windows”, “Wivik” and others) are well suited in combination with eye track, and they have been used in type-to-talk systems for decades. Character sets can be arranged in a traditional QWERTY order, in alphabetic order or according to frequency-of-use. Some systems offer predictions of the most likely next characters and words or give access to a dictionary of words related to a specific context such as “dinnertime”. While it’s unlikely that a gaze controlled communication system will ever achieve 56

First International Conference on Technology-based Learning with Disability communication rates comparable to unimpeded speech, which may be more than 150 words per minute for a fast-talking person, the long term goal is to achieve an input rate that is comparable to the QWERTY keyboard for typing, as this is usually sufficient to partake in on-line conversations (“chatting”). Present day eye gaze communication is efficient at a level that compares to text input methods on mobile devices (PDA and mobile phones), typically around 10 words per minute. One particular system, Dasher (Ward & Mackay, 2000), shows promise for expert users reaching the level of Morse code and cursive handwriting, typically around 25 words per minute. Some of the more expensive systems have sufficient precision to allow control of a mousepointer in a Windows-environment. The screen resolution may have to be at a rather low level (e.g. 600 x 800 pixels), and larger-than-normal icon size may have to be applied. A zooming principle that works with normal size icons and at a higher resolution is also provided by some of the manufactures. Advanced gaze tracking systems usually incorporate dwell-time clicking as a method for selection of icons, menus and links. Dwell selection works by looking steady at an object or letter for a preset time. Most users start with a dwell time setting at more than one second, but some experienced users can use a dwell time setting that is less than half a second. Most users like to have longer dwell times when they are tired in order to avoid unintended selections. There are only a few reports on gaze control of wheelchairs at this time; this is an area of research and development that needs further work. At this point in time gaze cannot be reliably controlled in a dynamic environment where, for instance, other people walk around. While it can be annoying to do an unintended selection on a personal computer, it may cause a serious accident if it happens while controlling a wheelchair. THE NEED FOR A DURABLE SWITCH Gaze interaction appears to be one of the most desirable alternatives for individuals who cannot direct cursor movement with their hands. Pointing the eyes is a natural action, it takes seemingly no effort, and the visual system along with some facial motor control, often remain functional when degenerative diseases or traumatic injuries affect most other motor systems. Once a cursor has been directed over an item for selection, some action, equivalent to a mouse click, must be made to complete the selection process. Most gaze communication systems incorporate a dwell function for simulating a simple click. Some systems also include the ability to detect an eye blink as a means of creating a simple click. Both methods of clicking have inherent problems that can make eye gaze as a communication method more difficult to utilize. In the case of dwell, the user must hold their gaze in a pre-defined range of motion for a predefined duration of time. This results in a delay in responsiveness of the eye gaze system (Hansen et al., 2003). Another problem is the selection of false targets from holding ones gaze too long in the wrong location. Most users need to practice for a couple of hours before they master dwell selection (Itoh et. al., 2006).

57

First International Conference on Technology-based Learning with Disability To employ intentional eye blinks for clicking the user must close and hold closed at least one eye for a pre-defined length of time. This time must be longer than that of involuntary eye blinks to prevent false selections from involuntary eye blinks. Thus the use of eye blinks for selection can delay the responsiveness of the eye gaze system. Further, some users report that the muscle effort needed to blink becomes tiring with constant use. At times the portability of a user’s gaze communication system may present a problem. In some cases it is not possible to take the eye gaze system along in the car or to a hospital for example. There may be times when the user’s eyes are tired but the user still needs an easy way to communicate simple messages. There may be users who will eventually need some other form of communication access to substitute for a gaze pointer that they can no longer control. A male person with ALS described his concern about a potential gradual loss of his eye control in an email: “This is an important question, since many final stage People with ALS (PALS) experience that their eye muscles become weaker and weaker. This is contrary to what is written in most textbooks on ALS, but it is actually happening. I can mention that some Japanese PALS, who have survived 20 years using ventilator, can only stare straight out. Secondly, many final stage PALS take drugs (Scopoderm, Atropin etc) against saliva. Unfortunately, these drugs interfere with the sight leading to loss of precision and accurateness.” The use of a facial muscle controlled EMG switch for selection can provide a desirable alternative to dwell or eye blink selection. An EMG click can compensate for noise on the eye gaze tracker. An EMG click can be combined with a dwell click. EMG clicks can be strung together as sequences or held for predefined durations to have other meanings, like mode shift, double click etc.

EMG-SWITCHES MAY BE FASTER THAN A FINGER BUTON There are some indications that EMG-switches may also be attractive because they can become very fast. In a study performed with able-bodied participants by the USAF in which a forehead EMG switch was compared to a finger switch, response accuracy was found to be extremely high, approximately 98%, and reaction times fell between 180- 200 ms, a range considered to be the limit of simple reaction time. Several participants achieved 15-20% faster reaction times with the forehead EMG switch than with a manual switch (Nelson et. al., 1996). Surraka et.al. (2004) found indications that gaze pointing in combination with EMG-clicking would be faster than hand controlled mouse pointing and clicking for longer movements. Three student subjects at IT University of Copenhagen conducted an experiment to compare gaze typing on the GazeTalk system in combination with dwell-, mouse- or EMG-switch activation. They used a Tobii-1750 system to point with their gaze. Clicking was done by either dwell, a standard mouse or by the use of a Brainfinger system to control an EMG-switch by activation of the forehead muscle, i.e. corrugator supercilii. In average (across 60 sentences typed), mouse clicking produced the fastest typing (10,3 wpm, s.d. = 0,8), the EMG-switch came second (8,1 wpm, s.d. = 2,4), a dwell time setting at 500 ms produced 7,7 wpm, (s.d. = 0,7) while a dwell time at 1000 ms produced 5,9 wpm (s.d.= 0,6). The fastest session average (12,2 wpm) was made by a subject using the EMG-switch to click. (Bech et. al., 2005).

58


In another experiment, conducted at Wright State University in Ohio USA, 3 subjects typed 3 blocks of 10 sentences on the GazeTalk system in scanning mode (without gaze tracker or any other pointer), with a fast setting for the step-time (300 ms). They typed an average of 2,66 wpm (s.d.= 0,86) when they used a finger on the mouse button to click, they typed 3,17 wpm (s.d.=1,02) when they used a forehead mounted EMG-switch (Brainfingers system), activated by a forehead muscle (frowning) and they typed 4,25 wpm (s.d.= 1,80) when they used their jaw(bite)-muscle to activate the forehead EMG-switch (Brainfingers system). The keystroke per character was lower for jaw activations (KSPC= 0,98, s.d. = 0,28) than for finger activations (KSPC=1,34, s.d.=0,45). This is attributed to the fact that the subjects often clicked to late on a target button with their finger; while with jaw activations they did not. In another study, Surraka et.al. (2005), found that EMG-clicking by a smiling technique was significantly faster and less erroneous than EMG-clicking by a frowning technique (i.e. using corrugator supercilii). The above observations point to the need for more research in ways to optimize the measurement and use of facial EMG signals. FACIAL EMG SWITCHES AND ALS/MND The majority of people who currently use gaze communication equipment usually have a degenerative disease such as ALS/MND. In the case of ALS, motor neurons die, reducing the number of motor units and leaving muscle fibers that have lost their nerve supply. These orphaned fibers re-attach to other motor neurons. This results in a decrease of functional motor units with an increase in action potentials of these motor units (Stashuk, 2001). We hypothesize that a decrease in the number of motor units results in a perceived need to produce more effort to achieve a muscle contraction. We have observed that users with ALS approach control tasks with a tendency to over control and work from a high muscle tension level. This tendency should be taken into consideration when implementing an EMG switch algorithm. We hypothesize that it would be beneficial for individuals with ALS to operate at a lower muscle tension baseline level and command smaller muscle contractions to create a trigger. In this way less muscle fibers would be recruited, the overall response could be lower, the rise and fall could be faster, and the effort needed less. This would result in a faster trigger requiring less effort. We feel that EMG-switch software should take into consideration the above findings by facilitating operation at lower baseline levels and triggering at lower levels. Symptoms of ALS//MND include facial weakness and muscle cramps. If the facial muscles don't work then EMG switching is obviously not an option. Our experience with ALS/MND patients is that most of them will retain some weak facial muscle control for a relatively long period - along with some eye movement control. However, in the very final “locked-in” stage, where there may be no facial muscle activity left at all, blood-flow or beta/alpha switch control in a switch scanning mode are the last communication channels possible. We are currently investigating the possibilities for improving the beta/alpha EEG switch control with a new technique that shows promise. The technique incorporates mounting a Brainfingers

59

First International Conference on Technology-based Learning with Disability sensing package along the midline of the head. The sensors are placed along the top of the head. Initial findings indicate that the user can produce alpha/beta brainwave responses more easily at this location than at the forehead. Preliminary midline top of the head EEG switch response activity suggests that it might result in a responsive measurement system. Hitachi has developed a cerebral blood-flow switch that allows locked-in patients to produce a yes/no answer with an accuracy of 80 % in an average of 36 seconds. (http://edition.cnn.com/2005/TECH/09/28/mind.reading.device/). CONCLUSION Facial muscle activity, the electomyographic (EMG) signal, may be used for clicking to enhance a user’s gaze communication. If used correctly, the EMG signal can provide a fast click that would overcome the problems associated with using a dwell or eye blink click. The EMG click could also be used with an on-screen keyboard in a switch scanning mode when the portability of a user’s eye gaze system is a problem. When introducing EMG-switches to individuals with ALS, one should consider that the individual may have a tendency to over control and work from a high muscle tension level and that the EMG signal may contain excessive individual motor unit firing. The EMG switch should be designed to cope with user tension levels and high levels of motor unit firing. REFERENCES Bech, C., Dam-Andersen, M., Hansen, J. E. & Schifter, L. R. (2005) Cyberlink & Eyetracking – multimodale systemer som fremtidens computernavigation. (Student report in Danish, English title: Cyberlink & Eyetracking – the future of multimodal systems for computer navigation.), IT University of Copenhagen, 65 pages. Hansen, J. P., Hansen, D. W., Johansen, A. S., Itoh, K. and Mashino, S (2003). Command without a click: Dwell time typing by mouse and gaze selections. Proceeding of the 9th IFIP TC 13 International Conference on Human-Computer Interaction, pp 121-128. Itoh, K., Aoki, H., and Hansen, J. P. 2006. A comparative usability study of two Japanese gaze typing systems. In Proceedings of the 2006 Symposium on Eye Tracking Research & Applications (San Diego, California, March 27 - 29, 2006). ETRA '06. ACM Press, New York, NY, 59-66. DOI= http://doi.acm.org/10.1145/1117309.1117344 Majaranta, P., and Räihä, K. J. 2002. Twenty Years of Eye Typing: Systems and Design Issues. In Proceedings of the Symposium on ETRA 2002: Eye Tracking Research & Applications Symposium 2002, New Orleans, LA, 15–22. Nelson, W., Hettinger, L., Cunningham, J., Roe, M., Hass, M., Dennis, L., Pick, L., Junker, A., Berg, C. (1996) Brain-body-actuated control: assessment of an alternative control technology for virtual environments. Proceedings of the 1996 IMAGE CONFERENCE, 225-232 Surakka, V., Illi, M. & Isokoski, P. (2004) Gazing and Frowning As a New Technique for Human-Computer Interaction. ACM Transactions on Applied Perception, 1, 40-56. Surakka1, V. Isokoski1, P., Illi1, M. and Salminen, K. (2005): Is it better to gaze and frown or gaze and smile when controlling user interfaces? Proceedings of Universal Access in Human-Computer Interaction (UAHCI 2005), Las Vegas. LEA, cd-rom.

60

First International Conference on Technology-based Learning with Disability Stashuk, D. (2001): EMG signal decomposition: how can it be accomplished and used? Journal of Electromyography and Kinesiology 11 (2001) 151-173 Ward, D. J., and MacKay, D. J. C. 2002. Fast Hands-Free Writing by Gaze Direction. Nature 418, p. 838, August 22.

61


Computer Access Using Electrical Signals From The Forehead: The Cyberlink™ In Action Julio C. Mateo and Markus A. Feufel Wright State University Dayton, OH ABSTRACT Computers can substantially improve the quality of life for people with disabilities. However, for the potential of computers to be fulfilled, adequate input systems to enable computer access are crucial. Individuals with disabilities affecting manual control can use modified or alternative keyboards and mice, alternative input systems using output from effectors other than hands, or brain-computer interfaces directly linking brain activity to computer actions. The Cyberlink™ is a hands-free system that uses electrical signals from the forehead to control a computer. Research has shown that the Cyberlink™ can serve as an alternative input system for people with multiple disabilities and it is often recommended as a last resort when other alternative input system cannot meet the user’s needs. Advantages of the Cyberlink™ include its little reliance on head position, its clicking speed, and its therapeutic value. Disadvantages include its sensitivity to users’ overt emotional reactions and its dependence on not fully understood and difficult-tocontrol brain signals. Undergoing and planned research is expected to improve our understanding of brain signals and guide the development of new software applications and the improvement of the current ones, making the Cyberlink™ more usable and accessible and a more valuable tool for all potential users. INTRODUCTION Computers can substantially improve the quality of life for people with disabilities. For example, computers may provide non-verbal individuals with type-to-speech systems and allow blind users to hear written text using a screen reader. The ability to perform these actions (i.e., speak and read) with the help of a computer can greatly improve the independence and opportunities for social interaction of these individuals. However, for individuals with motor disabilities, these benefits will be thwarted if they do not have an easy and reliable way to access computers. Thus, adequate input systems to enable computer access for individuals with motor disabilities are crucial for the potential of computers to be fulfilled. Most commercially available computers use a keyboard and a mouse as input devices. However, individuals with motor disabilities (e.g., cerebral palsy) may have difficulty or may be unable to use traditional input systems (i.e., keyboard and mouse). Fortunately, there are numerous alternatives to the traditional input systems and individuals with motor disabilities can choose to use modified versions of the traditional input systems or alternative input systems (e.g., eye tracker) to access computers. 62

First International Conference on Technology-based Learning with Disability In this paper, we discuss the potential of the Cyberlink™, a hands-free system that uses electrical signals detected at the forehead, as an alternative input system for people with motor disabilities. Some of the research exploring the potential of the Cyberlink™ (e.g., Marler, 2004) is reviewed and some advantages and disadvantages of the Cyberlink™, when compared to other alternative input systems, are presented. We conclude the paper by describing some of the research we are currently performing using the Cyberlink™, as well as the research projects we plan to undertake in the future to improve the usability, accessibility, and general value of the system. COMPUTER ACCESS FOR INDIVIDUALS WITH MOTOR DISABILITIES Traditional input systems require individuals to use their hands in order to control the computer. Therefore, any disability that affects manual control will affect the ability of that individual to access computers. Technological aids available for individuals that have difficulty using traditional input systems can be divided into three categories, depending on the motor control available to the user: 1. Alternative or modified keyboards and pointing devices are designed to aid users who have difficulty using traditional keyboards or mice, but who have at least limited use of their hands. Examples of these technologies include: the StickyKeys, FilterKeys, and ToggleKeys options (available in the Accessibility section of every Windows 2000 or XP operating system); special keyboards that facilitate typing by limiting the amount of fine control needed (e.g., IntelliKeys®) or the amount of gross control needed (e.g., USB Mini®); and joysticks or trackballs sometimes used instead of mice to minimize the range of movement (Cook & Hussey, 2002). 2. Alternative input systems that use motor output from effectors other than hands. Example of these systems include those tracking eye or head position to control the cursor on the computer screen, speech-recognition systems that use voice input for typing and other computer functions, and switches that are activated with different body parts (e.g., mouth, chin, head). 3. Brain-computer interfaces (BCI) use sensors to detect brain signals (e.g., electroencephalogram, EEG) and use them as input signal for a computer. Brain-computer interfaces can be divided into three groups depending on how invasive they are: a. Non-invasive BCIs use sensors placed outside the skull to detect brain activity. b. Partially invasive BCIs use sensors placed under the skull on the surface of the cortex. c. Invasive BCIs use sensors placed within the cortex. Wolpaw et al. (2002) also distinguished between dependent and independent BCIs. The main difference between these two types is whether motor output of effectors is necessary to generate the detected brain activity. Thus, dependent BCIs measure brain activity resulting from effectors’ motor output whereas independent BCIs measure brain activity that does not depend on effectors’ motor output. Although the systems in each category require a different level of motor ability from its users, it is possible that a person with motor abilities adequate to use devices from a category (e.g., limited use of the hands) prefers using a system that does not use these abilities (e.g., a speechrecognition system). Similarly, when detecting motor output from alternative effectors becomes difficult (e.g., due to spastic movements), users may find it more efficient to use brain signals instead.

63

First International Conference on Technology-based Learning with Disability THE CYBERLINK™ SYSTEM The Cyberlink™ is an alternative input system developed by Junker (1997). Following the framework presented in the previous section, the Cyberlink™ could be categorized as a noninvasive BCI because it uses sensors placed on the skin to detect electrical signals. Although the Cyberlink™ is generally used as a dependent BCI, some studies (e.g., Doherty et al., 1999; Junker et al., 2001) have suggested that it can also be used as an independent BCI. The Cyberlink™ consists of three components: a headband, an interface box, and a software application (i.e., Brainfingers™). The headband contains three sensors that are placed on the user’s forehead and detect and amplify electrical activity at this site. The sensor in the center is placed directly above the nose and detects ground, while the other two sensors detect brain activity. A single “brain-body signal” is derived from these two inputs at the headband and sent to the interface box, which filters it into three different signals (i.e., frequency bands). The lowfrequency band (0.2 – 3 Hz, EOG) is highly responsive to eye movements, the mid-frequency band (0.5 – 45 Hz, EEG) responds to brain activity that is relatively independent from external movements, and the high-frequency band (70 – 1000 Hz, EMG) is highly responsive to facialmuscle activity. The software application which goes along with the Cyberlink™ system is called Brainfingers™ and it subdivides these three signals into a total of 11 fingers (i.e., frequency bands): 3 fingers in the EOG region, 7 fingers in the EEG region, and the EMG finger. Each of these fingers can be mapped to control one or multiple functions in the computer ranging from single and double clicking to one- or two-dimensional cursor movement). Brainfingers™ also includes a series of games (e.g., Pong and Labyrinth) that users can play to practice using the electrical signals produced at their forehead to control the computer. Many different combinations of these 11 fingers can be used for two-dimensional cursor control with clicking capability (i.e., as replacement for a mouse). For example, when users have control over their facial muscles, EMG is often used to control vertical movement of the cursor and often also clicking. This signal is usually the easiest to control for these individuals. Then, one of the EOG fingers can be used to control horizontal movements of the cursor. Users with no control over the EMG signal can sometimes use alternative fingers in the EOG and EEG region to control cursor movement (e.g. Doherty et al., 1999). Past research using the Cyberlink™ The Cyberlink™ was originally developed as an alternative input system for military use (e.g., Nelson et al., 1997). However, studies soon started to explore the feasibility of the Cyberlink™ system as an alternative input for individuals with motor disabilities. These feasibility studies (e.g., Junker et al., 2001) found that individuals with no documented means to access a computer could successfully input information to a computer using the Cyberlink™. However, these studies report that the system was not easy to control for most users, recommending it only as a last resort. Doherty and his colleagues (2002) pointed out that, even though the Cyberlink™ improved the access of some individuals to recreation (e.g., playing Pong), it was difficult to use it for communication. In their studies, they explored different interface designs to allow users to reliably say yes or no without clicking. Their research led to the development of guidelines for the design of maze-like interfaces for individuals who can control two-dimensional cursor movement with EOG and EEG but cannot reliably produce mouse clicks.

64

First International Conference on Technology-based Learning with Disability Another group of studies explored the usefulness of the Cyberlink™ for students with multiple disabilities (e.g., Marler, 2004). Marler found that the system was very helpful for these individuals and her experience led to the development of a training tutorial to aid practitioners determine the best combination of fingers for each individual (Marler, Junker, & Lustre, 2006). In addition, Marler reported improvements in attention span, motivation, and mental-age diagnoses (upgraded from 6-18 months to 4-5.5 years of age) after her students used the Cyberlink™ twice a week for a period of eight weeks,. The Cyberlink™ can also be used in combination with other input systems. For example, Junker and Hansen (2006) found that combining gaze input (for pointing) with the EMG signal (for clicking) led to faster and more accurate performance than using gaze-derived measures (e.g., blinking or dwelling) for clicking. In fact, Junker and Hansen pointed out that clicking with the Cyberlink™ is even faster than clicking using a manual control (e.g., mouse) and some gamers have used this faster EMG clicking to improve their gaming performance. Advantages and disadvantages of the Cyberlink™ system One of the advantages of the Cyberlink™ when compared to some of the other alternative input systems (e.g., eye or head trackers) is that it is less sensitive to spastic movements affecting head position (Marler, 2004). However, when using the EMG signal, overt emotional reactions resulting from the execution of successful actions (e.g., laughing) sometimes activate the Cyberlink™, resulting in involuntary clicks. When compared to invasive or partially invasive BCIs, the Cyberlink™ has the obvious advantage that no surgical procedure is necessary. When compared to other non-invasive BCIs, it has the advantage that it is more aesthetically pleasing and easier to attach than BCIs that place electrodes all around the skull (e.g., Wolpaw et al., 2002). The placement of the sensors on the forehead is particularly convenient because it is one of the last areas affected by degenerative diseases (e.g., amyotrophic lateral sclerosis, ALS). However, the more aesthetically pleasing design has the downside that the signal detected by the sensors is less rich than the signal detected by more complicated arrays. As mentioned in the previous subsection, clicking using EMG is faster than clicking using the hand (Junker & Hansen, 2006). This is clearly an advantage of the Cyberlink™ (if the user is able to use EMG signals), not only in comparison to other alternative input systems, but also in comparison with traditional input devices. Another advantage of the Cyberlink™, which is shared with other input devices that enable individuals to do things that were previously impossible, is its “therapeutic” value. That is, the positive effect that using the device has on the user’s cognitive abilities (e.g., increased attention span or estimated mental age), perceived control, and motivation. In addition, some studies (e.g., Doherty et al., 1999) have pointed out the diagnostic value of the Cyberlink™ in the case of individuals with locked-in syndrome. Even though the Cyberlink™ has some clear advantages over its alternatives, it is usually only recommended to users with disabilities as a last resort (e.g., Doherty et al., 2002). That is, after all other alternative input systems (except invasive and partially invasive BCIs, which are not used regularly due to their obvious risk) have failed to meet the user’s needs. One of the main criticisms and limitations of the Cyberlink™ is that its EEG signals are currently poorly understood and difficult to control and, as a consequence, the usefulness of the Cyberlink™ as an exclusively independent BCI (i.e., for individuals with no motor output) is limited. In addition, the Brainfingers™ software is not as reliable as it would be desirable and

65

First International Conference on Technology-based Learning with Disability necessary if the Cyberlink™ were to be used for daily use. However, some control is better than no control, rendering even major problems “acceptable” for users that have no other means to access a computer or exert an kind of control over their environment (Doherty et al., 2002). Undergoing and future research using the Cyberlink™ system Our lab is currently exploring the possibilities of the Cyberlink™ as an alternative input system for people with multiple disabilities. For example, we are currently studying its potential to help individuals with disabilities in the community by analyzing what obstacles they encounter when adopting Cyberlink™ as an alternative input system. We are using the experience derived from these case studies to design new interactive software programs that can be controlled using the Cyberlink™ and help to facilitate computer access and, ultimately, communication opportunities. Although the Cyberlink™ is described throughout this paper as a means to provide computer access, access as such is not always motivating to a person that has never used a computer before and, as a consequence, does not know what to do with it. Rather, we have found that the computer’s potential to facilitate social interaction is a strong motivating factor for individuals previously unable to communicate. We have further found that software applications that use computer access as a tool to enable social interaction are preferred and more successful than those applications merely providing control over the computer and its functions. In the future, we plan to combine the knowledge obtained from our case studies with more formal experimental studies in order to improve the Cyberlink™ system. For example, we plan to explore in more detail how the EEG signals change under different cognitive situations (e.g., mental arithmetic, pleasurable thoughts, relaxation) in order to create guidelines to improve and train voluntary control of EEG signals. As part of this effort, we also plan to explore more effective ways to use the EEG signal to control the computer. For example, one approach is to identify patterns across the activity of several fingers and map these patterns, instead of the activation levels of a single finger, to particular computer actions (e.g., move the cursor up). We plan to conduct a usability analysis of the Brainfingers™ software in order to provide useful recommendations for future software upgrades. Currently, the Brainfingers™ interface is mostly visual. Auditory feedback is only provided to the users when threshold values are exceeded, while visual displays continuously show the level of the brain signals. This design makes the Cyberlink™ difficult to use for individuals who are blind or have low vision. We plan to explore auditory displays that could be used to improve the accessibility of the Brainfingers™ software to people with visual impairments. Lastly, we plan to explore the possibility of using Cyberlink™ for environmental and wheelchair control. SUMMARY AND CONCLUSIONS Past and undergoing research suggests that the Cyberlink™ is a promising alternative input system to facilitate computer access for people with severe and/or multiple disabilities. This system has numerous advantages over other alternative input systems (e.g., fast clicks using EMG and less sensitivity to spastic head movements than other systems) but it also has some limitations (e.g., emotional reactions to the system, such as laughing, often trigger unexpected actions and brain signals are not fully understood). Understanding better how EEG signals from the Cyberlink™ can be used to control a computer could greatly improve the usefulness of this system for individuals with limited or no muscular control.

66

First International Conference on Technology-based Learning with Disability We expect current and future research to identify what type of difficulties current Cyberlink™ users experience in their everyday life and use this information in conjunction with more formal experiments to improve our understanding of how EEG signals can be used more effectively to control a computer. We expect to collaborate in the development of new interactive software applications to use with the Cyberlink™ that engage users in social interaction. As well, we hope to make the current Brainfingers™ software more usable and accessible for individuals who are blind or have low vision. REFERENCES Cook, A. M., & Hussey, S. M. (2002). Assistive technologies: Principles and practice (Second Edition). St. Louis, MO: Mosby, Inc. Doherty, E. P., Bloor, C., Rizzo, J., Berg, C., Engel, W., & Cockton, G. (1999). Cyberlink – An interface for quadriplegic, traumatic brain injured, and nonverbal persons. Proceedings of the 3rd International Cognitive Technology Conference, USA, 237-250. Doherty, E. P., Cockton, G., Bloor, C., Rizzo, J., Blondina, B., & Davis, B. (2002). Yes/no or maybe – further evaluation of an interface for brain-injured patients. Interacting with Computers, 14, 341-358. Junker, A. M. (1997). United States Patent 5,692,517. Junker, A. M., & Hansen, J. P. (2006). Gaze pointing and facial EMG clicking. Proceedings of COGAIN 2006: ‘Gazing into the Future,’ Italy, 42-45. Junker, A. M., Sudkamp, T., Eachus, T., Mikov, T., Wegner, J., Edmister, E., Livick. S., Heiman-Patterson, T., & Goren, M. (2001). Hands-free computer access for the severely disabled. Retrieved January 25, 2007 from http://www.brainfingers.com/research.htm Marler, D. M. (2004). Cyberlink: Computer access for persons identified with multiple disabilities. Unpublished master’s thesis, California State University, Northridge. Marler, D. M., Junker, A. M., & Lustre, B. (2006). Training tutorial. Retrieved May 1st, 2007 from http://www.brainfingers.com Nelson, W. T., Hettinger, L. J., Cunningham, J. A., Roe, M. M., Hass, M. W., & Dennis, L. B. (1997). Navigating through virtual flight environments using brain-body-actuated control. Proceedings of the IEEE Virtual Reality Annual International Symposium, USA, 30-37. Wolpaw, J. R., Birbaumer, N., McFarland, D. J., Pfurtscheller, G., & Vaughan, T. M. (2002). Brain-computer interfaces for communication and control. Clinical Neurophysiology, 113, 767-791.

67


IBIS: Intelligent Binary Information Selection Katherine L. McCreight N-Space Analysis, LLC Xenia, OH

ABSTRACT The need to access information on the world wide web has become an expected part of many educational, work, and personal activities. Web search typically requires the user to type search terms and subsequently navigate hyperlinks via keyboard or mouse controls. This creates an increasingly significant barrier for individuals who are unable to use keyboard or mouse input devices. Current middleware adaptations generally seek to replicate the actions of keyboard typing, using other input modes together with software aids such as predictive text and rowcolumn scanning. The improvement of context-sensitive search techniques now provides the potential for the development of a radically different search interface that is not dependent on keyboard simulation. N-Space Analysis is currently developing the Intelligent Binary Information Selection (IBIS) system, in which users navigate the web by indicating positive and negative responses to a sequence of pages. A discussion of this prototype system illustrates the potential of this binary evaluation methodology for web browsing and web-based information retrieval. INTRODUCTION The need to access information on the world wide web has become an expected part of many educational, work, and personal activities. Web search typically requires the user to type search terms and subsequently navigate hyperlinks via keyboard or mouse controls. This creates an increasingly significant barrier for individuals who are unable to use keyboard or mouse input devices. Current middleware adaptations generally seek to replicate the actions of keyboard typing, using other input modes such as predictive text, as in the Dasher program ("Inference Group," n.d.), and row-column scanning, as in, the DiscoverPro program ("DiscoverPro," n.d.). The WebFormator program ("WebFormator," n.d.) facilitates hyperlink navigation by compiling all the hyperlinks of a page in a sidebar. Meanwhile, hardware solutions such as gaze-tracking devices emulate mouse input. Adaptive software design has focused on re-creating the standard, keyboard-based mode of computer access. However, improvements in search techniques now provide the potential for the development of a radically different search interface, using binary input as the primary mode of interaction.

68


The IBIS system observes the user's simple positive and negative responses to a series of web pages, deduces the user's search objectives, and presents a sequence of ever-more appropriate pages in response. Currently in the software development phase, we hope to implement IBIS in a variety of hardware environments, including blink, sip, and switch interfaces. IBIS could also be used with a mouse-based interface, to reduce keyboarding requirements for users with repetitive stress injuries. Here, we report on the design issues associated with binary search. We review developments in search engine technology that make binary search feasible, outline the current IBIS program design, and discuss plans for future development and testing. DEVELOPMENTS IN SEARCH ENGINE TECHNOLOGY Two developments in search engine technology make binary search feasible: query-by-example, and binary evaluation. Both retail and academic search utilities have been developed that make use of query-byexample. In this method, instead of typing specific search terms, the user indicates objects which are similar to the desired target. In retail websites, we find "more like this" options, which present the shopper with similar items. These comparisons may rely on item titles, or on keywords assigned by the retailer. Academic research sites offer more sophisticated analysis of examples, so that the user may enter a paragraph or select an entire article in order to direct a search for similar items; see, for example, the National Academies Press Discovery Engine ("Help regarding the National Academies Discovery Engine," n.d.). In another form of query-by-example, the commercial search engine Semetric lets patent developers enter their patent descriptions in order to identify similar patents ("Engenium Semetric OEM," n.d.). These query-by-example technologies analyze entire texts using mathematical techniques such as Bayesian inferencing and Latent Semantic Indexing (Deerwester, et al., 1990). Term searching tends to return irrelevant information (a search for 'cougar' returns websites about both cats and cars), while also excluding useful information (a search for 'cougar' ignores websites that use the synonym, 'puma'). Textual analysis allows researchers to find related concepts, even in the absence of specific term matches. Binary evaluation interfaces have been introduced in some search applications. The web browsing utility StumbleUpon ("StumbleUpon," n.d.) uses a combination of binary evaluation and user communities to present the user with web pages that match the user's particular interests. The Aware search engine ("Aware Search," n.d.) starts with traditional search term input, but then asks users to assign positive or negative values to the search results, enabling the program to prioritize subsequent search results according to the user's preferences. The IBIS system extends these methods, using binary evaluation to drive a system in which entire web pages serve as example queries.

69


IBIS PROGRAM DESIGN Program Design IBIS is written as a Java application, which calls both the Mozilla Firefox web browser and the Alexa Web Search utility ("Amazon.com: Alexa Web Search," n.d.). For development purposes, IBIS takes binary input from keyboard, using the + and - characters; this simulates binary input from an adaptive device. The program begins by presenting the user with a series of web pages used as seedpages. Seedpages are opened in the Firefox browser, and the user is asked to provide binary evaluation of each page. As the user indicates positive and negative rankings for various pages, IBIS compiles the text content of these pages in two data structures. A salience (i.e., relevance) calculation identifies terms which are more common in the positive results than in the negative results. Terms found in both sets are discarded as uninformative. A high salience score indicates that a term is preferentially found in the positive set, and is therefore relevant to the user's search objective. IBIS then uses the top one, two, or three most salient terms to initiate a web search, using the Alexa Web Search utility. Alexa returns a summary of ten web pages which match that search request. These pages are evaluated according to the current salience rankings. Pages are ranked higher if they contain many salient terms, while pages with many undesirable terms are rejected. The page selection calculation considers the first 500 words displayed on the web page; the calculation also includes the page title and description, as entered in the hypertext mark up language (HTML) for that page. Title and description words are weighted more heavily than display text. The most highly valued page is presented to the user for evaluation, while other high-ranking pages are added to a page pool for possible future presentation. Note that the system does not need to find the best web page on each cycle; we only need to find some page that moves us closer to the user's objective. Each time the user evaluates a new page, IBIS recalculates the term salience scores. A timing variable allows the program to pay more attention to the most recent evaluations. This cycle of salience calculation based on approved pages, followed by searching on salient terms, approximates the human behavior of successive searching. When a human user does not find appropriate results with an initial query to a search engine, he or she typically reviews the results that are returned, attempting to find better terms with which to make subsequent searches. For example: In reviewing software designed for the disabled, an initial search by a human user for "disabled AND software" might be less than satisfactory (including, for example, results about viruses that disable software programs). The user may then identify the pages that are somewhat relevant, notice the frequent use of the phrase, "assistive technology," and adopt that phrase to create a more successful search. IBIS automates this process.

70

First International Conference on Technology-based Learning with Disability Program Walkthrough In a recent test run, IBIS was given three seedpages: http://www.gardenguides.com, http://mathworld.wolfram.com, and http://www.math.net. The tester evaluated the garden page positively, and the two math pages negatively. The IBIS program calculated salience scores for the words contained in the body of each web page, assigning the highest scores to those words found only in the garden page. The highest scoring word, Plants, was used to initiate a web search through the Alexa Web Search utility. Alexa returned summary information for ten web pages that matched this search term. A score was computed for each of these new pages, based on the previously established term salience scores. The top-ranking page, http://calphotos.berkeley.edu/flora/, was presented to the user for further evaluation. This is a page showing photographs of flowers. The tester gave this page a negative rating, and term saliencies were recalculated, this time yielding the word Garden as the basis for the next web search, and leading to exploration of a different set of web pages, starting with the page, http://www.garden.com/. A second test run on the same seedpages yielded different results when the tester assigned a positive score to the two math pages, and a negative score to the garden page. This time, the most salient word was Mathematics; a search on this term resulted in the selection and presentation of the page http://www.mathematics-online.org/. PLANS FOR DEVELOPMENT Plans for continued development of IBIS include the integration of a paragraph scanning utility to allow the user greater precision in directing searches, the introduction of other measures of page similarity, the use of multiple dimensions to characterize similarity, and the automatic detection of implicit goal shifts. Paragraph Scanning IBIS currently uses the entire web page when calculating term salience. Eventually, we would like to incorporate a paragraph scanning utility that would allow users to specify which particular sections of the page are useful to them. When the user evaluates a page as positive, the program would then provide a scan of the paragraphs, highlighting each in turn. A subsequent positive entry would identify a paragraph of particular interest, and the words of that paragraph would be weighted more heavily in determining the next set of search terms. An option could be offered to scan through the words in the paragraph as well, to specify specific search terms. Paragraph scanning would also allow the program to compile a research record for use offline. The research record would include, for each page approved by the user, the page address in hyperlink format, and the text of the paragraphs selected by the user during the evaluation. Other Measures of Page Similarity The current IBIS version uses text comparison as a way to measure page similarity. Eventually, we want the IBIS program to draw on a variety of similarity measures to identify the pages

71

First International Conference on Technology-based Learning with Disability which most closely match the user's objectives. Two promising similarity measures are hyperlink structure and community browsing behavior. Hyperlinks represent semantic relationships, as recorded by the authors of the web pages. We can thus use hyperlink structure to identify related pages. Community browsing behavior offers another way to measure page similarity: The order in which individuals access web pages, averaged over many individuals, can be assumed to indicate some relationship among those pages. Here, we may draw on information provided by the Alexa toolbar ("Alexa Web Search," n.d.), which annotates web pages with a list of other frequently visited pages. Dimensions of Similarity A page which is similar to the current page for one user may be irrelevant from the point of view of another user. For example, consider a person who is searching the web for information about a particular wildflower. Suppose the current page is an advertisement for a company which sells wildflower seeds and plants, and contains a description of the plant, bloodroot. The user makes a positive response, indicating a desire to see similar pages. For a user interested in purchasing bloodroot, another page which also mentions that plant name and prices would be useful. But a user who is researching plant toxicity might want a page which is similar in having the name of the plant, but dissimilar in not containing pricing information. We can better accommodate individual similarity evaluations if we model similarity in multiple dimensions. Each page can be modeled as occupying a location in the various dimensions of content, function (retail site, news site, personal journal or blog, etc.), language, authority, and so on. Initially, therefore, there are several locations which could count as similar to a seed page. If we can determine which dimensions are important to the user, we can choose pages which are similar in those dimensions. Seedpages would be chosen to maximize the differentiation of dimensions. Implicit Goal Shifts As a user interacts with IBIS, the system keeps a record of previous choices, using that information to improve each selection. However, we need to prevent the system from locking in on one set of search terms. The program needs a certain amount of "wobble" in order to identify useful pages. We see web search as a process of creating and refining a mental model of available information; as the user narrows in on a target, search behavior may change in subtle ways. Initially, a researcher may seek pages on a general topic, and finds many pages that constitute useful results. As the researcher proceeds, he or she may shift search objectives to take advantage of new information gleaned during the search process. At this point, pages addressing the general topic are no longer as valuable. IBIS therefore incorporates a timing variable, in order to give more weight to recent evaluations when determining search terms. We are also considering the use of intention-inferencing algorithms, which observe user behavior and track changing objectives. These would be derived

72

First International Conference on Technology-based Learning with Disability from our previous work on modeling concepts and decision-making (Eggleston & McCreight, 2002). PLANS FOR USER TESTING Our prototype program shows that the IBIS process is able to select appropriate, novel pages for web search. Further refinement is required before testing, but we plan to test the system against keyboard-based search using the commercial search engine, Alexa ("Alexa the Web Information Company," n.d.). We are concerned with both accuracy and ease of use. Conventional measures of search engine accuracy such as number of false hits (irrelevant results) will have to be adapted somewhat, since IBIS operates by iterating through a series of false hits as it approximates the search target. A more appropriate measure of accuracy might be, Can IBIS retrieve the desired information within a specified number of iterations? Ease of use can be compared by measuring the number of actions (binary input and/or keystrokes) required to locate the target information. N-Space Analysis is currently seeking partners with experience in adaptive technology to help establish user tests. OTHER APPLICATIONS OF THE IBIS TECHNOLOGY Binary search methodology offers possibilities for improving search functions for local documents, for commercial web sites, and for mobile devices. We plan to develop IBIS as an add-on program that an individual user would run on his or her own computer. In addition to directing web searches, the IBIS program could be adapted to support local search of user-created documents, as well as to provide improved navigation of electronic course materials such as e-books. The IBIS software could also be used by web site designers, providing a binary search interface for their visitors. The inclusion of a binary search interface would help developers meet standards of accessibility in web site design ("Web-based Intranet and Internet Information and Applications," 2001). An IBIS interface could aid researchers working outside of their native languages, since searches can be conducted on the basis of page similarity, without the necessity of typing words in an unfamiliar language. This could improve automated access to information for immigrant populations, travelers, and academic researchers. The simplicity of the binary IBIS interface would make it useful for mobile devices, for which keyboard entry is awkward and time-consuming, while the similarity-based IBIS design is well suited for typical mobile browsing uses such as reviewing news reports and locating nearby restaurants.

73

First International Conference on Technology-based Learning with Disability CONCLUSION The IBIS program illustrates the potential of binary evaluation for web browsing and web search. With additional development, we hope to create a search utility that can be incorporated into a variety of search domains. We believe the development of binary search functions, in conjunction with other input methods, will better serve the needs of both disabled populations and the general public. REFERENCES Alexa the Web Information Company. (n.d.) Retrieved May 1, 2007, from http://www.alexa.com. Alexa Web Search - Quick Tour (n.d.) Retrieved May 1, 2007, from http://www.alexa.com/site/help/quicktour. Amazon.com: Alexa Web Search: Amazon Web Services. (n.d.) Retrieved May 1, 2007, from http://www.amazon.com/gp/browse.html?node=269962011. Aware Search | The World's First Teachable Search Tool! (n.d.) Retrieved May 1, 2007, from http://www.awaresearch.com/. Deerwester, S., Dumais, S. T., Furnas, G. W., Landauer, T. K., & Harshman, R. (1990). Indexing by latent semantic analysis. Journal of the American Society for Information Science, 41(6), 391407. Retrieved April 29, 2007 from http://citeseer.ist.psu.edu/deerwester90indexing.html. DiscoverPro. (n.d.) Retrieved May 1, 2007, from http://www.madentec.com/products/discoverpro.php. Eggleston, R.G., & McCreight, K. L. (2002) Concept learning: Knowing and reasoning in the DCOG architecture. In W. Gray & C. Schunn (Eds.), Proceedings of the Twenty-Fourth Annual Conference of the Cognitive Science Society. Fairfax, VA, August 7-10, 2002. Engenium Semetric OEM. (n.d.) Retrieved May 1, 2007, from http://www.engenium.com/products/oem.php. Help regarding the National Academies Discovery Engine (n.d.) Retrieved May 1, 2007, from http://lab.nap.edu/nap-cgi/discover.cgi?act=help. Inference Group: Dasher Project: Home (n.d.) Retrieved May 1, 2007, from http://www.inference.phy.cam.ac.uk/dasher/. StumbleUpon >> Welcome to StumbleUpon. (n.d.) Retrieved May 1, 2007, from http://www.stumbleupon.com. Web-based Intranet and Internet Information and Applications (1194.22). (2001)

74

First International Conference on Technology-based Learning with Disability Retrieved May 1, 2007, from http://www.access-board.gov/sec508/guide/1194.22.htm. WebFormator-Download, Step 1 of 2. (n.d.) Retrieved May 1, 2007, from http://www.webformator.com/englisch/index.php.

75


Enhancing Learning with Haptic Devices as a Possible Assistive Aid D. W. Repperger Air Force Research Laboratory AFRL/HECP, WPAFB, Ohio 45433 C. A. Phillips Wright State University, Department of Biomedical, Industrial, and Human Factors Engineering, Dayton, Ohio 45435 ABSTRACT A powerful modality that may help assist learning or the interaction of people with computers or other devices is haptics. The term “haptics” is a general expression to describe both afferent and efferent flows of forces and information to the human (Repperger & Phillips, 2006). The first experimental study reported here involved USAF pilots (Repperger, 1991) and suggested that haptics may provide some benefits to the physically challenged. The second investigation dealt with physically challenged subjects with spasticity (Repperger, Phillips, & Chelette, 1995a, 1995b, Chelette, Repperger, & Phillips, 1995). These concepts were then again related to military applications (Repperger, et al. 1997) to study how pilots could better deal with wind turbulence using virtual reality simulators (Repperger, et al. 2003). Finally, subliminal haptics demonstrated improved situational awareness (Repperger, 2004, and Repperger, Phillips, et al. 2005) using multiple vector force fields.

INTRODUCTION New technologies to assist learning with disability includes the application of haptics technology to a wide range of applications. The origin and modern use of haptics is only about 25 years old, but numerous applications have shown means of improving the sense of presence of an operator when he/she has to work in remote environments. The term “haptics” can apply not only to the sense of touch, but also to other interactions that occur as devices with external mechanical forces can act on a human to produce some positive interaction in terms of performance or cognitive benefit through an improved sense of presence or possibly reduced workload. In (Repperger, Phillips, & Chelette 1995a, 1995b, Chelette, Repperger, & Phillips, 1995) , a haptics study was conducted with four types of patients with physical challenges: spinal cord injury, stroke survivors, cerebral palsy, and traumatic brain injury. A specially designed control stick was developed to mediate some of the untoward neuromotor responses elicited from the subjects, through the design of distinctive force-reflection algorithms on the haptic

76

First International Conference on Technology-based Learning with Disability manipulandum. It was found that certain force-reflection algorithms could mitigate the effect of uncontrolled spasms, but, in addition, provide a paradigm for enabling control. One case in point was a young 16 year old boy who had become traumatic brain injured as a result of an automobile accident when he was 4 years old. For a period of 12 years after the accident, this subject had little voluntary motor control of his arms and hand. Thus he could not use a device such as a computer or keyboard. After several months of assessment and testing with the haptic manipulandum, this teenager was then able, with the use of a force reflecting joystick, to perform simple tracking tasks and could be then fitted to the use of a computer, since he could now control a cursor on a screen. This experience of enabling the physically challenged individuals showed the potential this technology can bring to the rehabilitation community. Another individual in the same study was a recent stroke survivor (CVA for cardio vascular accident). This subject was about 50 years old and employed as an electrical engineer prior to the occurrence of the CVA. Through the participation of this second person in the experiment, his ability to handle more complex tasks was ameliorated by his experience with the haptics technology. This was clearly demonstrated through rapid improvement in the coordination between motor actions and visual changes of the task on the monitor. It was as if the haptic stick was reteaching the brain proper actions or new paths to complete a visual tracking task and developing a cognitive bridge between the sensing modalities involving motor control and vision. In a later publication (Repperger et al. 1997), a survey was made of four different paradigms in which the haptics technology was shown to develop other benefits. These remunerations also include acceleration of learning (reduction of time to complete a task) as well as improvement of the final tracking performance levels reached. Pilots flying simulated aircraft in virtual environments were involved in these experiments in which the haptic stick could be used to improve disturbance rejection to exogenous noise sources, such as wind turbulence (Repperger, et al. 2003). The problem of a pilot flying an aircraft when subjected to external noise disturbances, is not unlike the case of the spastic patient in (Repperger, Phillips, & Chelette, 1995a,b) who must deal with internal noise disturbances (spasticity). A variety of force reflection regimes were tested to examine the efficacy of the force-reflecting stick controller to deal with the uncertainty issues of the external environment. In modern times, the application of haptics to learning and rehabilitation is now quite persuasive (Broren et al., 2002, Mikropoulos & Stroubouis, 2004, Holden, 2005, and Viirre & Ellisman, 2003) . One very common medical application involves laparoscopic surgery, which is now very widespread. Haptic devices can be used to train young medical doctors on the use of laparoscopic instruments by providing the physician the sense of feel or touch as the instrument encounters obstacles or changes in tissue, bone, or other types of viscera encountered. By using more of the haptics technology, the medical doctor can improve upon his skills in a simulation environment at the same time as shortening the time to become proficient. Simulated training of this type is now more prevalent before doctors actually work on live patients. Four studies are now described within the Air Force Research Laboratory which started initially with pilots in the USAF and then transferred to subjects with physical challenges and then back again to the military application. After the studies with the physically challenged patients were better understood, then the concepts were again brought back to the military using the knowledge gained from the prior work.

77

First International Conference on Technology-based Learning with Disability A COMPENDIUM OF RELATED APPLICATIONS Since brevity must be the style here, we report on studies where some benefit was achieved . Study 1 was the initial USAF haptic application with pilots. STUDY 1 – DISCOVERING BASIC RULES OF HAPTICS APPLICATIONS In Figure 1 is a block diagram of the experiment (Repperger, 1991) and Figure 2 shows the performance results. Seven pilots were involved in the Figure 1 configuration. In Figure 2, the most interesting performance result was that the minimum tracking error occurred when the stick output was most highly correlated with the tracking target input. This led to the concept that the pilot does best when he “sees the task in his hand.” Study [1] Force

Operator

Target Input

Display

Forcing Function

error signal e(t)

+

x(t) = S(t) +

P1(s) =

fT(t)

Temporal

Reflecting Joystick

Human

plant output

Plant Dynamics

Force Loop P2(s) = Impedance

Position tracking loop

Figure 3 – Experimental Paradigm to Investigate “Matched Impedance”

Fig. 1 Study 1 – Block Diagram Description Tracking error signal

fT=Target +

-

Human Operator

Joystick Output = S(t) Stick Component Orthogonal to fT

S(t)

Display

Study [1] Temporal

Stick Component Correlated to fT

From Plant Output

Figure 4a – Block Diagram Description Related to the Stick Orthogonality Issue Bad Performance

73%

Root Mean Square

37.3

Percent of

Tracking Error

Stick Correlation with fT

55% Percent of Stick

29.5

Output Correlated

Error Units

Stick Mismatched To Plant Dynamics

With Target fT(t) Good Performance

Stick Mismatched To Plant Dynamics

Stick Matched To Plant Dynamics

Figure 4b – Tracking Performance Results Dependent on The Stick Orthogonality Relationship

Fig. 2 – Performance Data from Study 1 Another way to characterize Figures 1-2 is to state, “What you see is what you feel” and is an excellent rule of thumb to use with the design of haptic devices. The second study took the

78

First International Conference on Technology-based Learning with Disability technology developed within the military and generalized it to physically challenged individuals with a grant with the Department of Veteran Affairs. STUDY 2 – SPASTICITY STUDY AND HAPTICS From the knowledge gained of experiment 1, the goal was to take a broader view and bring this concept to patients with physical challenges (Repperger, Phillips & Chelette 1995a,b, Chelette, Repperger, & Phillips 1995). Figure 3 displays a block diagram description of the task of placing a cursor cross inside a small box on a visual display. This was a Fitts’ Law paradigm as shown in Figure 4. Figure 5 is a rendering of how data are plotted on a Fitts’ Law paradigm. From Figure 5, information capacity can be determined from the reciprocal slope of the performance curve. Figure 6 illustrates the calculated capacity data from the physically challenged subjects. What was most interesting was that (with haptic feedback) some of the physically challenged individuals actually performed better than the normal controls used as a comparison when the controls did not enjoy the benefit of haptic feedback. Force Reflecting Joystick

Study [2] Spatial

Study [2] Spatial Square Box Target

Display

Target Input ft

Plant Output

Human Plant

Operator

Width = W1

fo

Dynamics

fo

W2

W2

Force Loop Force Reflection Algorithm

Cursor at time t0 Cursor at time tf

Amplitude = A

Position Loop Figure 6 – Display Task – Which Gives Rise to Fitts’ Law Figure 5– The Closed Loop System to Test Physically Challenged Individuals

Fig. 3 – Study 2 – Acquisition Task

Fig. 4 – Fitts’ Law Scenario

Study [2] Spatial Capacity in Bits/Second for Physically Challenged Group

Study [2] Spatial

Slow

4

Slope = seconds/bits

Seconds

3.5 3 2.5

Time to Complete Task

Average Capacity in Bits/Second

Fast

2

Series1

1.5 1

RT

Easy Task

Difficult Task 0.5 0

0

Displacement

Bits of Difficulty

Figure 7 – Fitts’ Law of Information Processing with Humans

Fig. 5 – Fitts’ Law to Calculate Capacity

Linear

Quadratic

Cubic

Stick Condition

Figure 8– Average Capacity Measures for the Four Stick Conditions for the Physically Challenged Group

Fig. 6 – Capacities Calculated

In Figure 6, there were significantly different capacities obtained when the force reflection algorithm was changed. The rationale for the improved performance is derived in an information-theoretic manner for this performance metric. It was observed that for very nonlinear haptic feedback algorithms, humans become more efficient because they act more like a force controller rather than a position controller. A force controller is much more efficient

79

First International Conference on Technology-based Learning with Disability because the time delay in processing information (through a spinal reflex arc) takes only about 70 milliseconds, but for the position loop (involving the vision system and the brain) it takes over 160 milliseconds. With reduced latency the overall performance (and capacity) are improved. The next step was then to take this concept back to military applications and bring it within the context of a virtual reality simulator. STUDY 3 – PILOTS AND REJECTION OF WIND TURBULANCE Figure 7 shows a tight landing task developed in a virtual reality simulator as illustrated in Figure 8 (Repperger, et al., 2003). Figure 9 portrays the haptic force reflection algorithm which was the best algorithm from study 2 (the most nonlinear haptic algorithm). Figure 10 displays the performance results for a variety of dependent measures gleaned from the tight landing task with and without the haptic feedback. Study [3] Spatial- VR

Fig. 11a 3a - Subject in CAVE Environment

Study [3] Spatial- VR Roll Error at Touchdown

Runway F-16

Actual Landing Point at Touchdown

Optimum Landing Pitch Error at Touchdown

z(t)

Point at Touchdown

Variable x(t) = Lateral Landing Error

y(t) = Longitudinal Landing Error

Fig. 11b 3b - Subject in Tracking Scenario

Vertical Velocity/Acceleration at Touchdown

Haptic Stick

Runway

Thrustmaster

Figure Fig.10 7 - Performance Metrics Considered at Landing

Fig. 7 – Tight Landing Task

Fig. 8 – Subject in VR Simulator Study [3] Spatial- VR

Normalized F-16 Landing Performance Variables

Force Reflected in Pounds Against the Operator’s Hand

Study [3] Spatial- VR

1.4 1.312 1.2

Normalized Percent

0.91 pounds

Haptics Off

1

0.8

Haptics On

0.604

0.6 0.843 0.4

0.141

0.496

0o

S2

0.2

0 -30o

0.130

30o

Orientation Angle Between Flight Path of Aircraft and Runway

FigureFig. 12 8 - Haptic Stick Force Reflection Algorithm

0.127

0

percent crashes (1 unit = 100%)

X-RMS (1 unit = 46 feet)

0.152

Y-RMS (1 unit=150 feet)

Dependent Performance Measures

Fig. 9 – Force Reflection Rule

S1 Z-velocity (1 unit = 9 ft/sec.)

S1=Haptics On, S2=Haptics Off

Figure Figure 139

Fig. 10 – Performance Data Obtained

The value added (in a performance sense) for the haptics condition was significant for many of the important quantifications of performance as related to making a tight landing. This includes continuous data such as RMS (root mean square) errors in the X, Y, and Z axes as well as event driven measures, such as percent of crashes. A crash was defined, as per the standards given in the USAF and other air agencies. 80


STUDY 4 – IMPROVING SA VIA SUBLIMINAL HAPTIC STIMULATION Figure 11 shows the final study discussed herein (Repperger, Phillips, et al., 2005). The goal was to use haptic feedback to improve the situational awareness (SA) about a visual tracking task but the level of haptic stimulation (two collinear vectors) would be presented in a subliminal manner. The haptic stimulation was conducted in a subtheshold manner but the overall performance improved for an optimum amount of haptic injection of stimulation. Figure 12 portrays the net performance results that occurred showing that an optimum amount of haptic stimulation is beneficial for performance, even though it is basically presented in a subliminal manner. The performance measure (y axis) in Figure 12 is the time the target cursor is outside of a specified window in Figure 11. Thus small amounts of this error reflect upon good tracking performance. The optimum haptic condition is the fourth data point to the right which has the lowest mean absolute error as well as the lowest standard deviation value of the data when averaged over subjects. This is quite typical of results from the haptic studies whereas when the haptic condition is beneficial not only does the mean error become minimized, but also the variability of these data are usually the lowest. This seems to be related to the subjects having an improved condition of situational awareness.

Display

Haptic Lateral Chair

Stick

Movement

Motion Chair to Produce Acceleration Fig. 11 – Motion Simulator for Haptics in Study 4

81


Error 14

Timeoff Target in Seconds

12

10

8

6

4

2

0 C0S0

C0S1

C1S2

C1S1

C2S2

Experimental Condition

Averages Across nine subjects - Time off Target Tracking

Fig. 12 – Performance vs. Haptic Conditions in Study 4 DISCUSSION AND CONCLUSIONS There is strong evidence to suggest that a number of haptic studies presented here provide performance benefits to healthy or physically challenged individuals. The results are certainly task specific and predicated on the particular physical challenge being considered. Nonlinear haptic force reflection algorithms seem to offer the most utility in the overall human-machine performance. REFERENCES Broren, J., Bjorkdahl, A., Pascher, R. & Rydmark, M. (2002). Virtual reality and haptics as an assessment device in the postacute phase after stroke. CyberPsychology & Behavior, 5(3), 207-211. Chelette, T. L., Repperger, D. W., & Phillips, C. A. (1995). Enhanced metrics for identification of forearm rehabilitation improvement. IEEE Transactions on Rehabilitation Engineering, 3(1), 122-131 Mikropoulos, T. A. & Stroubouis, V. (2004). Factors that influence presence in educational virtual environments. CyberPsychology & Behavior, October, 7(5),582-591. Holden, M. K. (2005). Virtual environments for motor rehabilitation: review. CyberPsychology & Behavior. 8(5), 187-211. Repperger, D. W. (1991). Active force reflection devices in teleoperation. IEEE Control Systems Magazine, 11(1), 52-56. Repperger, D. W., Phillips, C.A. & Chelette, T. L. (1995a). Study of spatially induced 'virtual force' with an information theoretic investigation of human performance. IEEE Transactions on Systems, Man, and Cybernetics, 25, No. 10, pp. 1392-1404. 82

First International Conference on Technology-based Learning with Disability Repperger, D. W., Phillips, C. A. & Chelette, T. L. (1995b). Performance study involving a force reflecting joystick for spastic individuals performing two types of tracking tasks. Perceptual and Motor Skills, 81(2), 561-562. Repperger, D. W., Haas, M. W., Brickman, B. J. Hettinger, L. J. Lu, L. & Roe, M.M., (1997). Design of a haptic stick interface as a pilot’s assistant in a high turbulence task environment. Perceptual and Motor Skills, 85 (3 Pt 2), 11391154. Repperger, D. W., Gilkey, R. H., Green, R., LaFleur, T. &Haas, M. W. (2003). The effects of haptic feedback and turbulence on landing performance using an immersive cave automatic virtual environment. Perceptual and Motor Skills, 97, 820-832. Repperger, D. W. (2004). Adaptive displays and controllers using alternative feedback. CyberPsychology & Behavior, 7(6), 645-652. Repperger, D. W., Phillips, C. A., Berlin, J. , Neidhard-Doll, A. & Haas, M. (2005). Human-machine haptic interface design using stochastic resonance methods. IEEE Transactions on Systems, Man, and Cybernetics, Part A, Humans and Systems, 35 (4), 574-582. Repperger, D. W. & Phillips, C. A. (2006). A haptics study involving physically challenged individuals. Encyclopedia of Biomedical Engineering, Wiley, 3, 17741780. Viirre, E. & Ellisman, M. (2003). Vertigo in virtual reality with haptics: case report. CyberPsychology & Behavior, 6(4), 429-431.

83


Delivering on the Promise of Plato’s Academy: Accessible STEM Curricula for the Universitas Scholarium of the 21st Century Michele G. Wheatly Forouzan Golshani Jeffrey Vernooy Wright State University. Dayton, OH

ABSTRACT This paper reviews the philosophy of a university education as well as the progress American universities have made towards achieving the ideals of Plato’s Academy. We then present a case study for Wright State University, a doctoral research intensive university that has developed niche programming in learning with disability. A commitment to construction of an accessible campus 40 years ago, has provided the foundation for accessible programming. In addition to provision of student services, WSU has established a culture of disability acceptance that has evolved into leadership in the education of students with disabilities at all levels from K-20. In this paper we summarize the landmark developments in a program that is layered at all levels in the educational pipeline and encompasses the mission of the academy including instruction, research and service. In the year that WSU celebrates its 40th anniversary, this paper will document efforts to “make a difference in the lives of people with disabilities”. It also serves as the starting point for discoveries and applications that will further enhance accessibility in the universitas scholarium of the 21st century. INTRODUCTION: A BRIEF HISTORY OF UNIVERSAL ACCESSIBILITY IN AMERICAN UNIVERSITIES Universities, recognized as institutions of higher learning, have been in existence possibly since 21st century BC in China and ancient India (Brockliss, 1991). The earliest, historically documented “university” was Plato’s Academy (387 BC). The first university offering a range of degrees was Al-Azhar in Cairo (10th century). Universities were established, by definition, to be “communities of scholars” intent on acquiring tertiary and quarternary education. While the philosophical intent of higher education was clearly to be inclusive of all thinkers (Republic VI, 456, 474), throughout history, access to corporate learning has paralleled cultural norms. The earliest American universities, like their European counterparts, were deeply rooted in the classical Western tradition emanating from Greece. The Jeffersonian model of a higher education prepared “gentlemen” for citizenship in a democracy. The 19th century witnessed the founding of the land grant universities (Morrill Act, 1862) in response to the agricultural and

84

First International Conference on Technology-based Learning with Disability industrial needs of a growing nation. The Research Universities were subsequently founded based on the Germanic model. Access to American universities burgeoned in the 20th century due the Serviceman’s Readjustment Act (“the GI bill” 1944) that opened up higher education to the populace on an unprecedented scale. Regional universities sprung up and, with funding from newly established federal agencies (NSF and NIH), a vast engine of research and development fuelled American economic prosperity post WWII. While the “universitas scholarium” was beginning to promote coeducation and opportunity for underprivileged students, minimal gains were made for students from minority populations, and women in Science Technology Engineering and Mathematics (STEM) fields. The situation for the “last minority”, people with disabilities, was not much improved, largely because of physical infrastructure on most college campuses as well as prevalent attitudes and expectations (Seymour and Hunter, 1998). Persons with disabilities have continued to struggle to achieve equity in all areas of their lives, but especially in education which translates to employability. Up until the Education of All Handicapped Children Act (EHA, PL-142, 1975) many students with disabilities were routinely excluded from public school. Sequential reauthorization of that legislation (first as IDEA, now as IDEIA, Individuals with Disabilities Educational Impact Act) mandated “inclusion”, namely that students with disabilities should be provided with the least restrictive public education alongside their peers. There is an extensive literature on the resources and training needed to appropriately prepare science and math educators for inclusion classrooms (Burgstahler and Nourse, 2000). In-service educators report lack of pre-service training for teaching special needs students as well as a low level of comfort with students with physical disabilities (Norman et al., 1998). As we enter the 21st century knowledge-based economy requiring “thinkers” rather than “manufacturing or service workers”, there are limitless opportunities to engage the cognitive skills of individuals with physical disabilities. In fact a recent report “Rising Above the Gathering Storm” (National Academies, 2007) recommends that America’s participation in the global economy will require opening up the educational pipeline for diverse learners including women, minorities and people with disabilities. The confluence of the Nation’s need for more “thinkers”, and emerging enabling technologies for people with disabilities promises to level the playing field for people with physical disabilities. Universities will play a key role in further opening access to this “last minority”. However, in order to do so, in the same way that Women’s Studies and Cultural Studies programs at our nation’s universities have enabled us to view our educational curriculum and delivery through the lens of women and minorities, viewing curriculum content and pedagogy through the lens of disability will be required. CASE STUDY: Wright State University This paper will present a case study of the pivotal role that a comprehensive research intensive university can play in emerging a reputation for creating universally accessible STEM curricula at the undergraduate and doctoral levels, strengthening the pipeline of students with physical disabilities who are prepared and motivated to pursue STEM degrees and careers. Accessibility on Campus Wright State University, founded in 1967, was constructed to be architecturally barrier free and, in its brief 40 year history, has emerged a culture of disability acceptance originating from the

85

First International Conference on Technology-based Learning with Disability on-campus presence of students, faculty and staff with physical disabilities. The Office of Disability Services established in 1970 (currently directed by Jeffrey Vernooy) provides the following services annually to 500 students (250 others function without need of personalized services) and 4500 students with disabilities have graduated. • • • •

Physical Support Services including personal assistance with daily living activities, mobility for students with multiple impairments, and complete access to all areas of the campus via a highly equipped grid of underground tunnels. Academic Support Services providing comprehensive assistance for students with disabilities, including sign language interpreters, note-takers, and test-proctoring scribes. Technology Center providing textbooks and classroom materials in alternative formats that include audiocassette tapes, computer disks, Braille, and tactual image enhancement. Career and Vocational Support Services providing assistance with career planning and development, opportunities for work site experiences, assistance with realistic career decisions and awareness on meeting the demands of chosen occupation.

Together, the barrier-free architecture and accompanying services for students with disabilities have provided a platform for faculty to consider developing academic programs dealing with the educational needs of the people with disabilities (rehabilitation, special education) but more importantly to reengineer the curriculum and delivery formats of all programs so that they are universally accessible. The on-campus presence of large numbers of individuals with disabilities has created a unique environment where able-bodied students are able to “look past the disability” and develop a level of disability acceptance unparalleled at other universities. In exit polls of graduating students, many able-bodied students comment that one of the most important lessons they learned on campus was to understand and accept disability. Accessibility in the Undergraduate Curriculum Over the past 10 years Wright State University has gained national recognition for accessible programming in STEM disciplines (Slack and Wheatly, 2004). Underrepresentation of individuals with disabilities has been especially profound in the STEM fields, a culmination of historical emphasis on “hands-on” approaches in the experiential sciences as well as the mathematical preparedness of entering students and the general perception of parents and other “gatekeepers” that these are not good employment fields for people with disabilities. In the early years WSU began offering adaptive labs for students with disabilities who were required to complete a year of general education lab science. This became the foundation, in the mid 1990s, for the NSF-funded Creating Laboratory Access for Science Students initiative (the CLASS Project, directed by Michele Wheatly) that established the “gold standard” for inclusive science education at the introductory level (Wheatly et al., 1999; Bargerhuff and Wheatly, 2004; Kirch et al., 2005, 2007; Lunsford and Bargerhuff, 2006). The CLASS team, a collaboration of faculty and staff from the College of Science and Mathematics, the College of Education and Human Services, and the Office of Disabilities Services, has undertaken a range of activities to enhance the success of students with physical disabilities in STEM fields including: Residential CLASS Summer Workshops for Science Educators Grades 7-16: To promote training in disability awareness and universal design of lab and field experiences using inquiry-

86

First International Conference on Technology-based Learning with Disability based science activities, workshop participants work directly with high school students with profound physical disabilities attending a summer science camp. Pre/post assessment of participants was performed using the “Teaching Science for Students with Disabilities Survey” instrument (Stefanich and Norman, 1996). Significant gains were made during the workshop in terms of preparedness and willingness to deliver science using universally designed constructivist approaches (Kirch et al., 2005). Over the life of the project approximately 50 educators have attended these workshops. Educators reflected on their experiences through journaling (Kirch et al., 2007) and confronted their own disability schemas. For example, after working with “Jerome” a student with left-side paralysis who uses a chair, “Jim”, a science educator, wrote: “In the afternoon, I worked with Jerome – I realized (powerfully) how much I still ruled by appearances and expectations based on them. I am not proud to admit this!!! I never expected Jerome to be as bright and engaging and (how unexpected) to have such a great sense of humor. Why did this surprise me so much? And if I have such prejudices—what about the rest of the world? I thought I was enlightened and open minded. I was as full of fear and misinformation as many other people. How humbling!”

Over the course of the workshop, in addition to a heightened level of comfort with disability, educators acknowledged their increased expectations of the students. Kirch and coworkers (2007) concluded that the primary benefit of direct interaction of educators and students with disabilities at the workshop was reinforcement of the “student as valued learner”. One wrote: “What I learned the most and was the most amazed with was the capability of the students to do the labs—their physical ability to do—especially with the help of the adapted equipment.”

Residential CLASS Science Camps for Grade 7-12 Students with Profound Physical Disabilities: To encourage students to consider a STEM career and to inform educator participants about the needs of a particular disability, students with a range of profound physical disabilities (approximately 30) attended a one-week residential camp. The CLASS team developed understanding of all aspects of offering residential educational programs for this student population. Students learned that they can live independently and most of these students subsequently elected to attend WSU. Educator Minigrants Program: To provide continued momentum for workshop participants to implement change at their home institution, a minigrant program was established (total of 7 grants of $5,000, impacting 700 students with disabilities). A sample of funded projects follows: • • • •

Several applicants purchased specialized equipment (wheelchair-accessible computer cart, SMART board, Verneir Software), re-wrote curriculum, and provided in-service training for their peers. One applicant used funds to revamp an unused area of the school grounds to design and build an accessible garden/greenhouse. One applicant used funds for construction of models for visually impaired students One educator funded a science “overnight” for 6/7 graders (30% had a disability)

87

First International Conference on Technology-based Learning with Disability Distance Learning Course: To reach a broader audience of educators, a distance learning course was developed and offered through Wright State University. Topics covered in this course included fair versus equal education, assistive technology for the science classroom, National and State Science Education Standards and how they relate to students with disabilities. The textbooks used for this course were edited by Stefanich (2001a,b). CLASS Showcased at Educator Conferences: CLASS materials and training were distributed at regional and national gatherings of educators including: the National Science Teacher’s Association, the National Association of Research in Science Teaching, the Geological Society of America, the Science Education Council of Ohio, CSUN Conference on Technology for Students with Disabilities, the Council for Exceptional Children, the Ohio Council of Teachers of Mathematics, and the National Association of Biology Teachers. In addition to making formal presentations and training attendees with video vignettes, CLASS often staffed a booth for wider distribution of the disseminable products that enabled gathering of additional survey data. ORCLASS: To broadcast project initiatives to a national audience, the CLASS project has just completed a pilot project in collaboration with the Ohio Resource Center, a virtual resource housed at the Eisenhower Center of Ohio State University (www.ohiorc.org). This is a webbased archive of science and math lessons that have already been peer-reviewed for meeting national and Ohio academic content standards. The CLASS project hired abstractor teams (content and special education specialist dyads) to draft accommodations for the pre-approved lessons, so that an educator, working under pressure to adapt a lesson for a given NSES standard, can easily acquire proven accommodations for a given disability. This pilot was so successful that ORC subsequently contracted with the CLASS project to provide accessibility to lessons in the reading domain. Accessibility as a Doctoral Concentration Recently a doctoral program has been developed at WSU to provide a broad and comprehensive education, realistic work experiences, and opportunities for problem-centered research in the area of Learning With Disability (LWD). An interdisciplinary team of faculty has collaborated to train a unique cohort of graduate students capable of bridging the gaps between the three main areas of: a) biology of disability, b) assistive technologies, and c) the pedagogy of individualized learning. The program is currently offered as an interdisciplinary concentration within four of WSU’s existing doctoral programs namely: Engineering (Eng), Computer Science and Engineering (CSE), Biomedical Sciences (BMS), and Human Factors/Industrial Organizational Psychology (HF/IO). The LWD doctoral concentration (directed by Forouzan Golshani) was approved in 2005 and was launched in January of 2006 with funding from the NSF’s Integrative Graduate Education and Research Traineeship. This First International Conference on Technology-based Learning with Disability is a direct outcome of the LWD IGERT. A key characteristic of the program is its emphasis on exploring the opportunities afforded by advanced technologies to expand human capabilities through multimodal interfaces and enhanced visualizations, and addressing the practical problems of the design of humantechnology systems that broaden and enhance learning so that it is universally accessible. The strength of this program is the framing of questions about the biology and nature of basic human

88

First International Conference on Technology-based Learning with Disability capabilities and limitations as well as about the potentials of modern technologies in a way that addresses concerns relevant to the design of effective systems for learning and development. The scholarly thrust consists of three interrelated research efforts in the areas of: i.

The basic nature of human performance (as viewed through the lens of disability): including basic molecular and physiological processes associated with human abilities and disabilities; basic computational processes in terms of information processing and control; and functional assessment of human performance and skill development.

ii.

The study of human-machine interactions (to inform assistive technologies): humanmachine interaction including language technology, multi-modal interfaces, virtual environments/collaboration technology, and prosthetic (replacement) or orthotic (assistive) sensory and motor technology.

iii.

Pedagogy (theory behind the science of learning that will transform training-systems development and universal access to learning): including the benefit that different technologies and instructional modalities will confer upon students with disabilities and other students who learn more effectively in that mode.

The intention of the program is that each student’s doctoral research project will study, design, or create experiences for people with disabilities to overcome obstacles or barriers in their lives and learning. While these programs reflect different methodological and theoretical traditions, there are many potential points of overlapping interest and intersection that favor an interdisciplinary approach. For example, there is broad interest in aspects of visual perception. These interests include work on basic models of macular degeneration (BMS), psychological studies of human motion perception (HF/IO), computational models of depth perception (CSE), and the development of prosthetic devices (Eng), all leading to research on the design of pedagogies for teaching STEM students with visual impairments. Students’ learning outcomes are shaped by a curriculum that combines a thorough disciplinespecific training with interdisciplinary studies and research experience in assistive methods, tools and technologies. To ensure that all graduates have the opportunity to develop a deep understanding of all dimensions of disability, its consequences, and possible remedies, the curriculum includes core coursework and a mandatory practicum at one of the partnering organization. The programmatic requirements are reported elsewhere (Golshani et al., 2007). Designing solutions for individuals with disabilities creates a healthy tension between the “basic” and “applied” STEM disciplines. The practical problem of designing more effective educational systems provides an important “test” for basic theories of human disabilities, human performance, information processing, learning, and pedagogy. This helps keep basic research from locking into paradigmatic assumptions that are inconsistent with the realities of life. However, the dialogue between basic and applied scientists also can help those attracted to the challenge of practical problems gain deeper appreciation for the value of theory and sound methodologies. These basic theories can sometimes be the most effective bridges for generalizations across disciplines and across the basic/applied gulf. Our hope is that the opportunity to challenge theory against the demands of designing practical solutions that enhance human potential may attract some people to graduate education who otherwise would not be interested. The emphasis on solving a pressing societal problem may be particularly attractive to persons from minority groups and women, as well as to students with disabilities.

89


VISION FOR THE 21 CENTURY: UNIVERSITIES FOR ALL SCHOLARS As WSU celebrates 40 years of accessibility in higher education, we pause to reflect on how best to position our programming to further enhance the educational success and the employability of individuals with disabilities. The collective wisdom of these three authors highlights the following areas of future work. Evaluation and early intervention: Universities like WSU, who are leading in providing facilities and education planning for students with disabilities, can serve the community at large through an Assessment Center where the individual needs of a student can be evaluated and appropriate interventions identified for implementation. Early intervention is key to career success. WSU proposes to establish such an Assessment Center for the state of Ohio, and which could service a broader geographical region accessible within a day’s drive. In addition to bringing students physically to the center, we have interest in communicating with a broader sector through webbased broadcasts. We also have interest in promoting interest in STEM through peer-mentoring programs, weekend enrichment programs and summer camp experiences. Strengthening the pipeline: Consistent with the national dialogue on STEM pipeline, there is added urgency to strengthen the pipeline for students with disabilities. At WSU anecdotal observations suggest that our successes with this population are focused on students who acquired their disability in high school or college but had previously been appropriately prepared for the rigor of STEM majors. In the future programs need to increase the competency and interest in STEM among students who are born with a disability. This will involve intentionally identifying students with aptitude for STEM earlier in their academic journey and working with their families and teachers to create multiple opportunities for them to experience success in STEM areas. We also propose to develop appropriate support to transition these students to higher education. Placing competent and caring teachers in our nation’s classrooms continues to challenge the implementation of “Science for All Americans”. As a leader in STEM education, WSU will continue to contribute to this national conversation on evidence-based research on learning science and mathematics. Power of partnership: In order to broaden the impact of WSU’s pioneering work in the area of making the STEM fields more accessible to individuals with disability, we intend to partner in intentional ways with other institutions/entities that have expertise in research, teaching and service that impacts the success of students with disabilities. Partnership will be key to the advancement of the educational needs and employability of individuals with disabilities. Partnership with other institutions of higher education is already underway specifically with University of Pittsburgh and Georgia Institute of Technology (conveniently accessible via the I70/I-75 corridors) primarily to leverage their research expertise to improve interventions for the large population of individuals with disability who work and learn daily at WSU. As WSU’s legacy transitions from a “place” for individuals with disabilities to a “force” that will serve as a change agent for individuals with disabilities, there is a need to partner with community agencies that support and enrich the daily lives of people with disabilities. Even more important is the need for the STEM researchers to engage in applications of their discoveries that can directly

90

First International Conference on Technology-based Learning with Disability benefit accessibility. As the population ages, accessibility will impact increasing numbers of Americans. REFERENCES Bargerhuff, M. E. & Wheatly, M. G. (2004). Teaching with CLASS: Creating Laboratory Access to Science for Students with disabilities. Teacher Education and Special Education, 27, 318-321. Brockliss, L. W. B. ed. (1991). History of universities. Oxford University Press: Oxford. Burgstahler, S. & Nourse, S. (2000). Accommodating students with disabilities in math and science. Seattle: WA, DO-IT: University of Washington. Golshani, F., Wheatly, M. G., Bargerhuff, M. E., Flach, J. & Vernooy, J. (2007). A multidisciplinary graduate program in technology-based learning with disability, in Proceedings 2007 ASEE Annual Conference & Exposition, Honolulu, Hawaii, June 2007 (in press). Kahn, S. (2002). Including all students in hands-on learning. ENC Focus, 10 (2), 14-17. Kirch, S. A., Bargerhuff, M. E., Cowan, H. J. & Wheatly, M. G. (2007). Reflections of educators in pursuit of inclusive school classrooms. Journal of Science Teacher Education (in press). Kirch, S. A., Bargerhuff, M. E., Turner, H. J. & Wheatly, M. G. (2005). Inclusive science education: classroom teacher and science educator experiences in CLASS workshops. School Science and Mathematics, 105, 1-22. Lunsford, S. K., & Bargerhuff, M. E. (2006). A project to make the lab more accessible to students with disabilities. Journal of Chemical Education, 83, 407-409. National Academies, (2007). Rising above the gathering storm. Energizing and employing America for a brighter economic future, The National Academies Press: DC. Norman, K., Caseau, D., & Stefanich, G. P. (1998). Teaching students with disabilities in inclusive science classrooms: Survey results. Science Education, 82, 127-146. Seymour, E., & Hunter, A. B. (1998). Talking about disability: the educational and work experiences of graduates and undergraduates with disability in science, mathematics and engineering majors. Washington, DC: AAAS. Stefanich, G. P. (2001a). Science Teaching in Inclusive Classrooms: Models & Applications. University of N. Iowa: Cedar Falls. Stefanich, G. P. (2001b). Teaching in Inclusive Classrooms: Theories & Foundations. University of N. Iowa: Cedar Falls. Stefanich, G. P., & Norman, K. I., (1996). Teaching science to students with disabilities: experiences and perceptions of classroom teachers and science educators. Association for the Education of Teachers in Science: DC. Slack, H. L., & Wheatly, M. G. (2004). Creating laboratory access for science students (CLASS): students and faculty perspectives in Invention and Impact: Building Excellence in Undergraduate STEM education for the diverse student population, AAAS: DC, 119123, AAS, DC. Available online at www.aaas.org/publications/books_reports/CCLI/PDFs/04_ILD_slack. pdf. Wheatly, M. G., Wood, T. J., Renick, P., & Vernooy, J. (1999). Making biology laboratories effective learning environments for students with disabilities: a national model for undergraduate instructors and grades 7-12 school teachers. Electronic Journal of Inclusive Education, 1 (3).

91


Connecting Teaching and Learning Greg P. Stefanich University of Northern Iowa ABSTRACT If American education is devoted to offering opportunities for all students to learn, then all teachers must have the knowledge to make appropriate adaptations so that every student, regardless of ability or disability, can become an active participant in the learning process. Effective teachers understand that what the student knows is really a construct within the mind of the learner which is mediated by circumstances unique to the individual. Teachers must make hundreds of decisions every day to learners who rely on them for guidance and support. Teachers who make a difference know their students well and make accommodations based on individual student needs. INTRODUCTION People with disabilities are from all societal classifications, yet more are disproportionately identified from disfranchised groups. The educational environment is an ideal setting to blend together persons with physical, cultural, emotional, and intellectual diversity into a cohesive, mutually supportive group. However, this can only be accomplished if classroom teachers have a knowledge and understanding of factors that affect learner engagement and strategies that improve participation. This paper contains a review of successful teaching practices in science with specific attention to sub-groups that are underrepresented in the fields of science, technology, engineering and mathematics (STEM). Petersen (2006) investigated the intersectionality of gender, race, disability, and class in American schools. She presents a powerful case that by having several stigmatizing identities that oppression is often exacerbated. Vernon (1999), describes that, “one plus one does not equal two oppressions” (p.385), they can be experienced simultaneously or singularly depending upon the context. Peterson (2006) argues that individuals lives cannot be understood through only one aspect of identity, albeit gender, race, disability or class. An understanding of one’s experience can be understood only through a thorough inquiry into the multiple dimensions of one’s identity. There is an increasing awareness that traditional science instruction favors white and Asian male children without disabilities. The majority of persons involved in science related professions that receive their education in American schools come from middle level or high socio-economic backgrounds and from English-speaking homes (Gibbons, 2003). Instructional congruence occurs when teachers mediate the nature of academic content and inquiry with consideration for language, cultural diversity, and disability (Lee and Fradd, 2001). The basic premise of

92

First International Conference on Technology-based Learning with Disability instructional congruence is centered on teacher behaviors and choices. The essence of the concept is to teach through the minds of the learners rather than expecting to learners to adapt to the thinking processes forwarded by the teacher. A common misconception held by many educators is that teaching and learning is a linear process, teachers teach and students learn. The fact is that, for most learners, the process is slow, convoluted, and further complicated by forgetfulness. Effective teachers have knowledge of their students and their needs. The following sections contain a meta-analysis of successful practice for each marginalized subgroup, followed by common elements that appear to improve learning for all students. RACE AND GENDER In recent years collaborative research supports the basic effect of school studies done in the late 1970s and early 1980s. In related publications Eggen (2002), Taylor, et al (2000), and Zeichner (1996) note characteristics of teachers who are able to produce relatively high levels of student achievement in culturally diverse settings. These are: ¾ Teachers have a clear sense of their own ethnic and cultural identities. ¾ Teachers are personally committed to achieving equity for all students and believe that they are capable of making a difference in their students' learning. ¾ Teachers develop a personal bond with their students in a democratic and cooperative learning atmosphere. ¾ Scaffolding is provided by teachers that link the academically challenging and inclusive curriculum to the cultural resources that students bring to school. ¾ Parents and community members are encouraged to become involved in students' education and are given a significant voice in making important school decisions in relation to program, i.e., sources and staffing. African American Students African Americans are an ethnic group within mainstream America with their own culture that may affect their performance in science in unique ways. There may be special challenges in connecting the home, community and classroom science learning. Teacher expectations, particularly for African American youth living in low socio-economic environments are lower than those held for other ethnic populations (Diamond, Randolph and Spillane, 2004). The danger exists in how teachers approach the learners. When students are approached as if they are less capable than their peers, they achieve less, are absent more often, and are more likely to drop out of school. There is a particular danger with grouping or tracking students. Tracking does not increase student performance but, in contrast, increases the potential for school failure and dropout (Pettit, 2006, p. 107). Msengi (2006), summarized that triangulating information between the adult family member, the student and the associated teacher can be a tool for a shared understanding leading to improved home-school relationships. He states further states that these efforts can build on family strengths and provide flexible venues for exploring alternatives rather than dwelling on conflicts between the adult family member, the student and the teacher. Anderson and Stokes (1984)

93

First International Conference on Technology-based Learning with Disability reported that Anglo families were much more likely to initiate educational interactions, where African American families waited for the child to initiate. Differences between the language of the home and school can vary in conversational behavior (turn taking between an adult and child when conversing), motivation (whether a child feels rewarded by either the family or teacher), and learning style (learning by observing before performing) (Reese, et al., 2000). In a study by Linek, Rasinski and Harkins (1997), over 90% of teachers recognized the importance of involving parents, yet less than 5% supported involving parents as partners. Teachers that do not make an effort to know the family are failing to know the student. In science classrooms where there were high proportions of students who were from African American background, students performed better and enjoyed school more when teachers: 1. Students achieved more when teachers provide hands-on experiential learning opportunities for students in science (Simpson, 2002). 2. Higher student outcomes were achieved when students were allowed to work together in cooperative groups (Hilberg and Tharp, 2002). 3. The availability of a mentor to help the teacher address circumstances unique to the cultural context of the community was beneficial to student learning (Starnes, 2006). 4. Improved student motivation and engagement was found when teachers walked through investigations on occasion to guide students through the thinking process. (Curtin, 2006). 5. There was improved student performance when teachers initiated family engagement and provided guidance to guardians about what adults could do to help (Sonnenschein & Schmidt, 2000). 6. The student-teacher relationship improved when teachers gave examples and not too many directions. (Curtin, 2006). 7. Exposure to key concepts and vocabulary was important early in the lesson. More effective teachers worked to bring in language connections and student experiences associated with the concepts (Krashen, 1994). 8. Time to practice particularly through games and exposure to multiple intelligence activities resulted in improved student engagement and improved student attitudes towards their teachers (Curtin, 2006). 9. Teachers who walked around the classroom, sought out student responses instead of waiting for students to raise their hands, and teachers who encouraged students were more effective (Starnes, 2006). American Indian Children Starnes (2006) notes that we do not recognize or respond to the chasm that exists between customary methods and curriculum and the way American Indian children learn. In addition the little we know about the history, culture, and communities in which they live comes from white educators. A special challenge exists because each culture and each community is unique. The hardships associated with efforts at extinction, slavery, forced migration, forced religious conversion and disease are all imbedded in historical roots. This are complicated with contemporary social challenges involving poverty, unemployment, teen pregnancy, substance abuse, higher than average dropout rates, and higher than average levels of suicide.

94

First International Conference on Technology-based Learning with Disability Specifically reflecting on her work with Native American youth, Starnes (2006) offers a number of suggestions for teachers. First and foremost is the need to seek out a mentor who will help guide decisions through the cultural and historical circumstances unique to the cultural context where you are teaching. Teachers need to attend appropriate cultural, social and sporting events so the community senses there is intent to be connected. When possible bring in responsible elders as speakers to students and in other roles where they can enhance school-community relationships and communication. Non-native teachers need to become educated abut history from the perspective of the community and examine difficult realities where one looks at injustices committed against people by people or by government policies and practices, both historical and contemporary. There is a need to create materials and activities specific to your curriculum because there is a general unlikelihood that commercial materials exist. Lastly one must be patient and expect different cultural issues to emerge again and again. One must understand that it takes time to change generations of cultural interactions and mistrust. In science classrooms where there were high proportions of students who were from American Indian background, students performed better and enjoyed school more when teachers: ¾ Employed hands-on, experiential learning in an informal, flexible learning environment (Gilliland, 1999 and Simpson, 2002). ¾ Employed collaborative processes where students were given opportunities to work together (Hilberg & Tharp, 2002). ¾ Presented the whole concept before focusing on segments and details (Cajete, 1999). ¾ Brought in visual aids or models whenever possible (Riding and Rayner, 1998). ¾ Had a mentor available to help guide teacher decisions through the cultural and historical circumstances unique to the cultural context (Starnes, 2006). ¾ Attended appropriate cultural, social, and sporting events (Starnes, 2006). English Language Learners For students who do not experience the English language as young children one can expect delays in their conceptual understanding of science. These delays exist even in students that appear to have good social verbal skills and effective communication with everyday vocabulary. One of the greatest challenges for teachers relating to English Language Learners is that of engagement. Without direct communication with the learner(s) the learning gaps that are likely to occur makes science an insurmountable hurdle. In an ethnographic study of a group of largely Hispanic students from Mexico, Curtin (2006) reported that the students liked school in the United States. They felt materially comfortable and safe, even though they attended an urban school in a low-socioeconomic section of a large city. One area of difficulty they noted, even though they felt fluent in English, was a struggle with the content and vocabulary in science. Students focused on faces and non-verbal communication from teachers. They noted frustration with assessment and grading practices utilized by their teachers. Curtin (2006) shared numerous suggestions from in-depth conversations with adolescent English Language Learners. They described good teachers as ones that used examples, explained a lot, and did not give too many directions. They spent a lot of time

95

First International Conference on Technology-based Learning with Disability focusing on their actual teaching strategies. They walked around the classroom and sought out students instead of waiting for students to raise their hands. In science classrooms where there were high proportions of students who were from English Language Learners, students performed better and enjoyed school more when teachers: ¾ Used hands-on approaches (Curtin, 2006) ¾ Walked through a problem rather than requiring students to figure it out on their own (Curtin, 2006). ¾ Provided the necessary supports for success and held high expectations (Curtin, 2006). ¾ Chunked information into short segments (10 minutes) and allowed short times to discuss with a neighbor what was taught (Hansen, 2006). ¾ Used projects, drawings, labeled diagrams, and posters in assessment (Hansen, 2006). ¾ Did not “blame” students for not learning and were always seeking new teaching strategies (Curtin, 2006). Gender Women indicate declining interests in entering STEM fields beginning in the 7th grade (VanLeuvan, 2004), and they are more likely to drop out of science majors after entering a college or university (Vidal-Arwin, 2002). Girls with disabilities face double discrimination. They are hit with the bias that girls can’t do math and science and that people with disabilities can’t do math and science. This is compounded by a perception that somehow their disability makes them weak, needy, incompetent, or dependent, which often translates into protecting them from challenging work and learning (Wahl, 2001). Girls have strengths that are from their talents and form a perspective somewhat different from their male counterparts. The following strategies, offered by Wahl, 2001, are ways to put these principles into practice. In science classrooms, girls performed better and enjoyed school more when teachers: ¾ ¾ ¾ ¾ ¾

Were non-sexist and strong leaders (Settles & Cortina, 2006). Used cooperative rather than competitive learning environments (Vidal, 2002). Demonstrated appreciation for women’s participation in science (Vidal, 2002). Served as mentors and guides (Downing & Crosby, 2005). Encouraged girls to ask questions and encouraged girls to persist in science investigations (Wahl, 2001). ¾ Encouraged girls to take intellectual risks; to make mistakes and try again in the quest for understanding; to get messy (Wahl, 2001). ¾ Supported girls to resist traditional socialization that values being neat, getting the right answer, and being compliant and unquestioning (Wahl, 2001).

Lesbian, Gay, Bisexual, Transgender, or Questioning It is important that educational environments accept the responsibility of all of the students in their care. One aspect of uniqueness that is unsettling for many, somewhat because it often transcends religious or moral convictions, is the issue of homosexuality. Persons with disabilities

96

First International Conference on Technology-based Learning with Disability who develop new feelings based upon their sexual awakening and physiological development can often feel that they are in love-feelings that can make them feel uncomfortable and confused, not knowing how to act or react. This becomes even more complex when associated with being identified as gay or lesbian, The responsibilities of teachers and educators cannot be ignored in this regard. In addition to issues of social adjustment and mental health there are significant issues regarding safety and wellness. While some educators may feel uncomfortable in addressing these issues based upon their own fears or beliefs, it is a professional responsibility of educators, in all positions and at all levels, to provide a safe and supportive environment for all students. In schools where there are initiatives relating to alleviating discriminatory actions towards students with disabilities there are significant differences. The average grade point of LGBT students is more than 10 percent higher, they are 40 percent less likely to skip school, and they are twice as likely to report that they intend to go to college. Equitable education ensures that everyone is valued and appreciated for the qualities they have without consequences by those that might be hypocritical, prejudiced or biased because of their own beliefs. In science classrooms, Lesbian, Gay, Bisexual or Transsexual students performed better and enjoyed school more when teachers: ¾ Completed professional training on LGBT needs, and skills in meeting those needs (Bailey N., 2005). ¾ Acted assertively to prevent harassment, demeaning statements, demeaning gestures, or isolating maneuvers by others (Bailey N., 2005). ¾ Became aware of a safe person to whom students can turn to get accurate information about sexual orientation or gender identity (Bailey N., 2005). ¾ Examined the school curriculum to look at ways to appropriately incorporate history, literature, and role models form the LGBT community to forward a message, just as we need to do with all other marginalized groups, that in America everyone can live a successful and productive life (Bailey, N., 2005). ¾ Examined the library and multi-media sources to ensure that there is an age appropriate body of information and literature on L, G, B, T (Bailey N., 2005). Students with Disabilities American education has a long-standing history of low expectations for students with disabilities. These expectations influence (a) the amount of time students with disabilities spend learning science and the amount of time they are actively engaged in investigation in the science laboratory, (b) the academic focus and quality of objectives, (c) the sequence and depth of the learning opportunities afforded to students, (d) access to knowledge and resources, (e) homework expectations, and (f) curriculum alignment. Common dangers students with disabilities encounter are the cumulative effects of sympathy and low expectations. A convenient teaching action is just to expect less from those who don't perform well on common group assignments. Similar low expectations occur by having the student with disabilities only be an observer during science experiments or by providing peer assistance in activities that require writing or fine motor coordination. When students realize low expectations, it may be the result of teacher

97

First International Conference on Technology-based Learning with Disability decisions that sustain mediocrity in students. On the other hand, expecting too much can be equally devastating. Many physical disabilities place a strain on a student’s endurance and time. The student probably needs more rest, may take longer to get ready for school or to get set up for homework, and may find activities physically challenging that are routine for most students. In many cases the student with disabilities does not get to handle things because of teacher inconvenience, unwillingness to make adaptations, or just an overall lack of awareness and sensitivity of classroom teachers. Early experiences, extending throughout the school-age years, often instill in disabled students the idea that their role is one of a "passive observer" in an active learning setting. As a result, students with disabilities often become passive observers in activity-centered settings because their unique needs are not considered with sufficient positive regard. In science classrooms, students with disabilities performed better and enjoyed school more when teachers: ¾ Provided multiple exposures to new terms and concepts (Wood, 1990). ¾ Adjusted assignments and acquired alternative resources (Lawrence, 1988). ¾ Allowed students opportunities for exploration using heterogeneous cooperative groups (Krashen, 1994). ¾ Used projects, drawings, labeled diagrams, and posters in assessment (Hansen, 2006). ¾ Made carefully designed adaptations in the general education setting, rather than relying on pull-out programs (Stainback, Stainback, and Stefanich, 1996). Successful Practice Highly effective teachers adapt and modify. One area of consistency is that when teachers assume responsibility to engage all students the achievement gap narrows and classroom achievement improves (Haycock, 2002). Significant differences can be seen between teachers who teach groups of students and those who look at the uniqueness of each individual. Communication and understanding are needed for all students, but especially for those with disabilities. Three essential C’s earmark the qualities and attributes of effective teachers. The first is credibility; teachers need a knowledge of their subject and the ability to communicate it to the learner in ways that the learner can equilibrate. A second essential quality is that of caring. Effective teachers have a way of making each and every student feel important and appreciated. Students need a school situation that treats them with dignity and respect and demonstrates caring about them. A third attribute of effective teachers is connectedness, the ability to mobilize other parties to support the learning process for the student. Teaming with other professionals is an essential skill in today’s schools. Under ideal situations there are parents/caregivers who value learning and who know how to provide practical support. In other cases the teacher must learn resiliency and adaptability in learning how to work with at-risk adults. In science classrooms, all students performed better and enjoyed school more when teachers:

98

First International Conference on Technology-based Learning with Disability ¾ Employed hands-on, experiential learning in an informal, flexible learning environment (Gilliland, 1999 and Simpson, 2002). ¾ Employed collaborative processes where students were given opportunities to work together (Hilberg & Tharp, 2002). ¾ Incorporated games (Curtin, 2006). ¾ Brought in visual aids or models whenever possible (Riding and Rayner, 1998). ¾ Provided the necessary supports for success and held high expectations (Curtin, 2006). Because of the special challenges of integration for people with disabilities, the educational environment is an ideal setting to blend together persons with physical, cultural, emotional, and intellectual diversity into a cohesive, mutually supportive group. Group synergism can become more powerful than the collective output of each person acting independently. In a democracy, this is what society is searching for and what businesses seek in their employees. Why shouldn't it be the essence of what we wish to accomplish in schooling? To accomplish these outcomes, the talents and abilities of everyone must be considered. The environment and interactions must be adjusted and modified to meet the capabilities and potentials of each individual. REFERENCES Bailey, N. J. (2005, November). Let us not forget to support LGBT youth in the middle school years. Middle School Journal, 37(2), 31-35. Cajete, G. (1999). Igniting the sparkle: An indigenous science education model. Skyland, NC: Kivaki Press. Curtin, E.M. (2006, January). Lessons on effective teaching from middle school ESL students. Middle School Journal 37(3), 38-45. Diamond, J.B., Randolph, A. & Spillane, J.P. (2004). Teachers’ expectations and sense of responsibility for student learning: The importance of race, culture, and organizational habitus. Anthropology and Educational Quarterly, 35(1), 75-98. Downing, R.A., & Crosby, F.J. (2005, December). The perceived importance of developmental relationships on women undergraduates’ pursuit of science. Psychology of Women Quarterly, 29(4), 419-426. Eggen, B. (2002, February). Administrative accountability and the novice teacher. Paper presented at the Annual Meeting of the American Association of Colleges for Teacher Education, New York City. (ERIC Document Reproduction Service No. ED 464 050). Gibbons, B. A. (2003). Supporting elementary science education for English learners: A constructivist evaluation instrument. The Journal of Educational Research, 96(6), 371-380. Gilliland, H. (1999). Teaching the Native American (4th ed). Dubuque, IA: Kendall Hunt.

Hansen, L. (2006, January). Strategies for ELL success. Science & Children 43(4), 22-25.

99

First International Conference on Technology-based Learning with Disability Haycock, K. (2002, December). Toward a fair distribution of teacher talent. Educational Leadership

60(4), 11-15. Hilberg, R. S., & Tharp, R. (2002). Theoretical perspectives, research findings, and classroom implications of the learning styles of American Indian and Alaska native students. ERIC Digest, ED, 468-000. Krashen, S. (1994) Bilingual education and second language acquisition theory. In C. F. Leyba (Ed.), Schooling and language minority students (pp. 61-63). Los Angeles, CA: California State University, Los Angeles. Lee, O., & Fradd, S. H. (2001). Instructional congruence to promote science learning and literacy development for linguistically diverse students. In D. R. Lavoie & W. M. Roth (Eds.), Models of science teacher preparation (pp. 109-126). AA Dordrecht, The Netherlands: Kluwer Academic Publishers. Linek, W. M., Rasinski, T. V., & Harkins, D. M. (1997). Teacher perceptions of parent involvement in literacy education. Reading Horizon, 38(2), 90-107. Msengi, S. G. (2006). Family, child, teacher perceptions of what African American adult family members think and do to assist their elementary school-aged children to become better readers. Unpublished doctoral dissertation, University of Northern Iowa, Cedar Falls. Petersen, A. (2006). Exploring intersectionality in education: The intersection of gender, race, disability, and class. Unpublished doctoral dissertation. University of Northern Iowa, Cedar Falls. Pettit, S. (2006). There are no winners here: Teacher thinking and student underachievement in the 6th grade. Unpublished doctoral dissertation, University of Northern Iowa, Cedar Falls. Reese, L., Kroesen, K., & Gallimore, C. (2000). Agency and school performance among urban Latino youth. In R. Taylor & M. Wang (Eds.), Resilience across contexts: Family, work, culture and community (pp. 295-332). Mahwah, NJ: Lawrence Erlbaum Associates. Riding, R.J., & Rayner, S. (Eds.). (1998). Cognitive styles and learning strategies: Understanding style differences in learning and behavior. London: David Fulton. Settles, I. H., Cortina, L. M., Malley, J., & Stewart, A. J. (2006, March). The climate for women in academic science: The good, the bad, and the changeable. Psychology of Women Quarterly, 30(1), 47-58. Simpson, L. (2002). Stories, dreams, and ceremonies: Anishnaabe ways of learning. Tribal College: Journal of American Indian Higher Education, 11(4), 26-29. Sonnenschein, S., & Schmidt, D. (2000). Fostering home and community connections to support children’s reading. In L. Baker, M. J. Dreher, & J.T. Guthrie (Eds.),

100

First International Conference on Technology-based Learning with Disability Engaging Young Readers: Promoting achievement and motivation (pp. 264-284). New York: Guilford. Stainback, W., Stainback, S., Stefanich, G.P., & Alper, S. (1996). Learning in inclusive classrooms: What about the curriculum? In S. Stainback and W. Stainback, (Eds.), Inclusion: A guide for educators (pp. 209-220). Baltimore, MD: Paul Brookes Publishing Company. Starnes, B. A. (2006). What we don’t know can hurt them: White teachers, Indian children. Phi Delta Kappan, 87(5), 384-392. Taylor, B. M., Pearson, P. D., Clark, K., & Walpole, S. (2000). Effective schools and accomplished teachers: Lessons about primary-grade reading instruction in low-income schools. The Elementary School Journal, 101(2), 121-166. VanLeuvan, P. (2004, May/June). Young women’s science/mathematics career goals from seventh grade to high school graduation. The Journal of Educational Research, 97(5), 248267. Vernon, A. (1999). The dialectics of multiple identities and the disabled people’s movement, 14(3), 385-398. Vidal, A. (2002). Methods for retention of undergraduate women in science majors. Geological Society of America, 34(6), 121. Wahl, E. (2001). Can she really do science? Gender disparities in math and science education. In H. Rousso & M.L. Wehmeyer (Eds.), Double jeopardy: Addressing gender equity in special education (pp. 133-153). Albany: State University of New York Press. Zeichner, K. (1996). Educating teachers for cultural diversity. In K. Zeichner, S. Melnick, & M.L. Gomez (Eds.), Currents of reform in preservice teacher education (pp. 133-175). New York: Teachers College Press.

101


Inclusive Science Education: Classroom Teacher and Science Educator Experiences in CLASS Workshops* Dr. Susan Kirch Queens College Dr. Mary Ellen Bargerhuff Wright State University Heidi Cowan Ohio State University ABSTRACT Inclusion is the meaningful participation of students with disabilities in general education classrooms. The CLASS project (Creating Laboratory Access for Science Students) is a unique initiative offering training and resources to help educators provide students with a variety of physical, sensory and learning disabilities equal access in the science laboratory or field. To determine whether participants believed a 2-week residential workshop sponsored by CLASS raised disability awareness and provided teacher training in inclusive science teaching practice, a likert scale survey and questionnaire was completed by all participants (N=20) in four workshops. Participants reported large gains in their familiarity with instructional strategies, curricula, and resources and their ability to design, select, and modify activities for students with disabilities. Finally, shifts in attitudes about teaching science to students with disabilities were noted. INTRODUCTION A substantial part of educational reform in the sciences is directed toward students who have been customarily excluded from the field of science (Salend, 1998). The National Committee on Science Education Standards and Assessment (1993) asserted, “The commitment to Science for All implies inclusion not only of those who traditionally have received encouragement and opportunity to pursue science, but for women and girls, all racial and ethnic groups, the physically and educationally challenged, and those with limited English proficiency” (p. 5). There are several barriers which can contribute to students with disabilities not pursuing science related fields. These barriers include historical, institutional, physical, attitudinal, and curricular barriers. The CLASS (Creating Laboratory Access for Science Students) Project uses professional development workshops to prepare educators for inclusive classrooms, thereby helping them eliminate barriers for their students. Students with disabilities have struggled for their educational rights for decades. The Education of All Handicapped Children Act (EHA) in 1975 encouraged students with disabilities to attend public schools. This legislation was reauthorized as the Individuals with Disabilities

102

First International Conference on Technology-based Learning with Disability Education Act (IDEA) and now calls for free and appropriate public education for all students in their least restrictive environment. Even though the definition of a child’s “least restrictive environment” is not clear, the law does emphasize the need for students to be educated alongside their typically developing peers. This educational setting is to be used unless education in the general education classes with the use of supplementary aides and services cannot be achieved satisfactorily (Hocutt, Martin, & McKinney, 1990). To investigate teachers’ attitudes toward mainstreaming and inclusion, Scruggs and Mastropieri (1996) conducted a meta-analysis of survey data from 28 studies spanning 37 years (1958-1995) and including 10,560 general education and special education teachers. They found that nearly two thirds of general education teachers surveyed supported the concept of mainstreaming/inclusion. More than half felt that mainstreaming/inclusion could provide some benefits, but less than one third believed that they had sufficient time, training, or resources necessary to implement it. Their study along with other similar research uncovered variables that seem to be related to successful inclusive science instruction (Mastropieri, Scruggs, & Bohs, 1994; Scruggs & Mastropieri, 1994). These variables included (a) administrative support, (b) support from special education personnel, (c) an accepting, positive classroom environment, (d) appropriate curriculum, (e) effective general teaching skills, (f) peer assistance, and (g) disability-specific teaching skills. Efforts to prepare teachers for inclusive classrooms have continued to expand since the passage of IDEA. Reiff, Evans, and Cass (1991) reported that 37 of the 50 states required general educators to take at least one university course in teaching students with disabilities. One course, however, cannot adequately prepare teachers for the complexity of inclusive classrooms. “There continues to be a need to provide inservice training to build the specialized competencies required for the inclusion of students with disabilities” (Johnson, 2000, p. 281). There are many resources available to help teachers and administrators enact inclusion in ways that benefit all students in classroom instruction. There are tips in the CLASS sourcebook for science teachers to help teachers design and implement effective learning in a laboratory or field environment (Wood, T.S., 2001). The sourcebook was written in collaboration with professionals from biology, science education, special education, rehabilitation, disability services, and psychology, with input from persons with physical and learning disabilities, parents, and others who were able to lend their expertise to the project. In addition to the CLASS sourcebook, there are several other individuals and organizations that have produced high quality resources for educators in the last 10 years. These books, brochures, and tapes typically include practical strategies for teaching individuals with disabilities in the classroom and school and techniques for overcoming barriers to inclusion. The Creating Laboratory Access for Science Students (CLASS) Summer Workshop was designed to provide educators with the opportunity to present science lab activities to a group of high school students with physical, sensory, or specific learning disabilities, thereby increasing their ability and willingness to teach in inclusive classrooms. A detailed description of all CLASS Project activities, including summer workshops, can be found in Bargerhuff and Wheatley (2004) and Bargerhuff, Wheatly, and Kirch (in press). Therefore, this paper presents only a summary of a typical 2-week residential workshop. The workshops discussed in this paper were offered in the summers of 1999, 2000, and 2001. The first week of the workshop was for educators only. Activities and topics covered during the first week included disability awareness and sensitivity training. Popular demonstration lab and field exercises that had been modified to accommodate students with

103

First International Conference on Technology-based Learning with Disability disabilities were introduced throughout the week. Educators were also introduced to a variety of adaptive technologies and computer programs that individuals with disabilities have found improve their access to information and communication. During the second week, upper middle school and high school students with disabilities (grades 7-12) joined the educators on campus for a week of Summer Science Camp. These students worked through science activities with the CLASS Summer Workshop participants. Educators worked one on one with a different student for each exercise. After each lab, educators met with a team of project staff (instructor, lab manager, and special educator) for a debriefing session on the inclusive aspects of the lab or field exercise. Through this type of “direct experience” workshop model, educators used popular science demonstration lessons that had been adapted to accommodate the needs of the student participants. Although there are several exceptional programs that work with teachers to modify and adapt science activities for students with learning disabilities or severe emotional disturbances (Caseau & Norman, 1997; Cawley et al., 2002; Finson, Ormsbee, Jensen, & Powers, 1997; Hammrich, Price, & Slesaransky-Poe, 2001; Moroney, Finson, Beaver, & Jensen, 2003; Ormsbee & Finson, 2000), the CLASS Project specifically used this direct experience workshop approach to provide professional development for teachers wanting to meet the needs of students with disabilities in their science classrooms. This direct experience workshop format is in keeping with our understanding of how educators gain pedagogical content knowledge through teaching (Cochran-Smith & Lytle, 1993). The guiding questions were the following: 1. Do teachers feel better prepared to teach science to students with disabilities after participating in the workshop? 2. Have teachers’ attitudes toward teaching science to students with a disability changed after participating in the workshop? METHODOLOGY Description of Sample The CLASS Advisory Committee selected workshop participants. The committee reviewed workshop participants’ written applications. Participants who were accepted into the program demonstrated a clear desire to broaden their knowledge of inclusive practices in addition to scientific concepts covered in the activities. The committee gave preference to those who had worked in inclusive environments in the past or those who were currently developing inclusive science classrooms for students with physical or learning disabilities. They paid particular attention to those applicants who were part of a collaborative special education/general education team from the same school, since these partners had greater potential to influence organizational change in their districts. The workshop participants (N = 20; 5-8/year) in this study had diverse backgrounds, including (a) science teachers who anticipated inclusion classrooms, (b) special educators who sought more training on how to teach science to students with disabilities, (c) staff at university/college learning centers who provided assistance to students with disabilities, and (d) instructors at major research universities involved in introductory labs. Participants represented 18 schools from 11 states and ranged in age from 28-50. In order to determine if the strategies used in the CLASS workshop actually addressed barriers faced by students with disabilities who are interested in science careers (such as teachers’ disability awareness and attitudes toward inclusion), a survey was conducted of 6 Inclusive Science Education workshop participants. Of

104

First International Conference on Technology-based Learning with Disability the participants surveyed, 90% reported having no training in their preservice program on how to accommodate students with specific disabilities in science. The CLASS workshop was designed to address this professional training need. Workshop Evaluation Method: Survey An anonymous, self-reporting survey was used to evaluate the impact of the workshop. In each of the 3 years (1999-2001), all participants engaged in pre- and postsurveys using the Teaching Science for Students With Disabilities instrument developed by members of the Association for the Education of Teachers in Science Committee for the Inclusion of Challenged Populations (Norman et al., 1998; Stefanich & Norman, 1996; Appendix A). This survey examines four categories, including the following: (a) exposure to issues regarding students with disabilities, (b) preparation for teaching science to students with disabilities, (c) strategies for teaching science to students with disabilities, and (d) attitudes about teaching science to students with disabilities. The survey consists of a series of questions on which participants were asked to respond on a multipoint Likert scale. The only nonparametric statistics appropriate for use on the survey results were the Chi-square statistic or the Kolmogorov-Smirnov (K-S) one-sample test. The sample size (N = 20) for this study was too small for the Chi-square test to be meaningful. The K-S one-sample test is an alternative test of goodness-of-fit useful for small sample sizes, but it requires comparing an observed cumulative frequency distribution with a theoretical distribution that cannot be determined for the preworkshop sample. At best, these data are ordinal, and there is a continuum underlying the observed scores, while the actual scores fell into discrete categories. Therefore, descriptive statistics were used to analyze this data set. Workshop Evaluation Method: Questionnaire Participants were also asked to complete a short questionnaire at the end of the workshop. They responded to six questions regarding various aspects of the workshop, including the following: 1. What was your favorite experience during the workshop? Why? 2. What was your least favorite experience? Why? 3. Please provide suggestions or commends on the laboratory activities. 4. Please provide suggestions or comments about group discussions. 5. What new information have you learned through the workshop that you plan to implement in your classroom/lab? 6. Do you have other concerns/suggestions about anything not covered in the other questions? Qualitative research methods were used for analysis of questionnaire responses. The first two authors read the responses to each question and analyzed data inductively by using categorical coding (Bogdan & Bilken, 2003). The data were categorized into broad areas, with quotes and examples that supported each category. The two researchers compared responses and resolved differences in coding, with agreements and disagreements recorded to check on intercoder reliability (as in Burstein, Sears, Wilcoxen, Cabello, & Spagna, 2004). The two researchers reorganized responses into general categories that fit under a single-phrase them specifying the general trend in their reflections (as in Hamre & Oyler, 2004). Three themes emerged from the statements made in response to the question, “What was your favorite workshop experience and why?” 1. Student as Valued Learner. 2. Knowledge of Self. 3. Physical Environment as Fundamental Support to Learning.

105

First International Conference on Technology-based Learning with Disability Three themes emerged from the statements made in response to “What new information have you learned through the workshop that you plan to implement in your classroom/lab?” 1. A Call to Change Personal Practice. 2. Enhancing Collegial Collaboration. 3. Educators Advocate. RESULTS Experience With Students With Disabilities The survey asked teachers to identify the disability categories corresponding to the types of disabilities of students they were currently teaching or have taught in the past. IDEA has defined 13 categories of students with disabilities, typically divided into high incidence and low incidence categories. When teachers who have participated in the CLASS workshops were surveyed, they reported trends similar to those reported by the U.S. Department of Education (2001). Preparation for Teaching Science to Students With Disabilities Different but related efforts are often necessary to include students with disabilities in the science lesson (Tomlinson, 1999). Teachers in our study related the coverage each of the disability categories received during their teacher education program in college, both in science methods and general teacher education classes (Table 1). More than half of all participants reported that they did not receive information about students in any disability category other than specific learning disability and Attention Deficit/Hyperactivity Disorder (ADHD). Table 1: Coverage Received in Teacher Preparation Courses Coverage Received in Teacher Preparation Courses IDEA Disability Category General Teacher Education Science Methods Courses* Courses* Specific learning disabilities 3 2 Speech or language 2 1 impairments Mental Retardation --Emotional Disturbance 2 1 Multiple disabilities 1 1 Hearing impairments 2 1 Orthopedic impairments 2 1 Physical health 2 1 ADHD 2 1 Visual impairments 2 1 Autism 2 1 Deaf-blindness 1 1 Traumatic brain injury 2 1 Developmental Delay 2 1 *Teacher participants were asked to “rate the coverage of each of the disability categories received during your teacher education program in college, both in science methods courses and general teacher education classes.” Ranking scale: 5=comprehensive coverage, 4=moderate coverage, 3=adequate coverage, 2= minimal coverage, and 1=no coverage. Mean responses were calculated for each category – general teacher education and science methods (N=20). The mean scores were rounded to the nearest whole number (e.g., 2.1-2.4=2 and 2.5-2.9=3).

106


Participants within the study reported feeling more comfortable with working with students with disabilities (Table 2). Table 2: Reported Preparedness for Teaching Science Pre- and Post-Workshop IDEA Disability Reported Preparedness for Teaching Science Category Pre-workshop* Post-workshop** Reported Change*** Specific learning 3 4 1 disabilities Speech or language 2 3 1 impairments Mental Retardation ---Emotional Disturbance 2 3 1 Multiple disabilities 2 3 1 Hearing impairments 2 3 1 Orthopedic impairments 2 4 2 Physical health 2 3 1 ADHD 3 3 0 Visual impairments 2 3 1 Autism 2 2 0 Deaf-blindness 1 3 2 Traumatic brain injury 2 3 1 Developmental Delay 3 3 0 *Before the workshop teacher participants were asked, “In the blank beside each of the categories of disabilities, please write the number which best represents your preparedness to teach science to students with the indicated disabilities.” Rating scale: 5=comprehensive coverage, 4=moderate coverage, 3=adequate coverage, 2= minimal coverage, and 1=no coverage. This is the mean response of the participants (N=20). **On the last day of the workshop teacher participants were again asked “In the blank beside each of the categories of disabilities, please write the number which best represents your preparedness to teach science to students with the indicated disabilities.” Rating scale: 5=comprehensive coverage, 4=moderate coverage, 3=adequate coverage, 2= minimal coverage, and 1=no coverage. This is the mean response of the participants (N=20). ***Reported change in preparedness=mean response (post) – mean response (pre).

Topic Areas Relevant to Working With Students With Disabilities Creating inclusive environments for students with disabilities requires that educators remove both physical and curricular barriers. Educators must be familiar with supports available for staff and students, effective instructional practices, and curricular adaptations (Lipsky & Gartner, 1997) in order to overcome these obstacles. When participants were asked to report their preparedness in these different topic areas, they reported the largest gains in their familiarity with resources that provided information on teaching science to students with disabilities. They noted gains in how to use research on best practice for teaching students with disabilities. They reported they had become familiar with suggestions that address student needs in different disability categories. Since the focus of the CLASS workshop was to teach educators about adaptations and modifications they could make to classroom equipment and common laboratory exercises, it is not surprising that many of them cited that this is what they learned.

107


Attitudes About Teaching Science to Students With Disabilities Individuals with disabilities face constant discrimination. Teachers in the CLASS workshop experienced an intensive 2-week program in which they were asked to rethink their attitudes and perceptions of individuals with disabilities in many ways. To determine whether participants experienced any attitudinal shifts, respondents were asked to agree or disagree with statements regarding educating students with disabilities before and after the workshop. Participants used a 4-point Likert scale (1 = strongly agree, 2 = agree, 3 = disagree, and 4 = strongly disagree) to indicate their level of agreement or disagreement with 30 statements. We do not have the data to draw confidently any conclusions between the preworkshop and postworkshop data, but we observed that there were more questions with significantly larger proportions in the postworkshop results than in the preworkshop results. There were noticeable changes in attitude indicated by the shift in responses of the participants to the following five statements. The implications of these shifts are described in the discussion section. • “Students with disabilities are at-risk in terms of safety in hands-on science.” Only 1 participant agreed with this statement at the end of the workshop, a change from 7 participants prior to the beginning of the workshop. • “I feel inadequate in my preparation for teaching science to a student with a physical disability.” Three participants agreed with this statement on the postworkshop survey compared to 10 on the preworkshop survey. • “Care should be taken to not give students with disabilities unrealistic goal expectations which will inevitably result in frustration when they try to find employment.” Nine participants agreed with this statement after the workshop; 14 agreed before the workshop. DISCUSSION The purpose of this study was to evaluate the impact a 2-week, direct experience, residential workshop had on participating educators. The educators in this study reported gains in every category surveyed. Presurvey results from this sample are in keeping with the findings from previous research (Norman et al., 1998). Educators feel they are inadequately prepared and trained regarding teaching science to students with disabilities. They realize they have limited knowledge about methods and adaptations for students with disabilities, and teachers are unclear about appropriate expectations, assessment, and grading for students with disabilities. After participation in the workshop, however, participants reported feeling better prepared for many instructional tasks in science. At the workshop, participants interacted primarily with students who had disabilities in low incidence categories (e.g., motor/orthopedic, hearing, visual, and multicategorical impairments). This direct experience with students could account for the large gains reported by educators for these specific disability categories. This finding is in contrast to the small improvements reported by teachers for the students in the other categories. This result may be taken as, “support-of-principle,” illustrating that providing one-to-one experiences with students with disabilities in a direct experience workshop is an effective way to build teacher confidence. Note, however, that this is not simply a “learn-by-doing” model (Bell, Blair, Crawford, & Lederman, 2003). We do not believe educators will automatically learn how to teach science to all individuals with disabilities just because they have an opportunity to work one on one with students. The inquiry-workshop experience required that educators also participate in disability awareness training and in explicit discussions with workshop trainers about how and why to use particular teaching strategies,

108

First International Conference on Technology-based Learning with Disability curricular adaptations, and technological supports. This conclusion is also supported by the change in the number of participants who agreed with the statement, “I feel inadequate in my preparation for teaching science to a student with a disability,” from 50% of participants before the workshop to only 15% afterwards. Interestingly, individuals participating in the workshop held many of the same negative attitudes or misconceptions that Norman et al. (1998) uncovered previously. Without preworkshop interviews, it is difficult to determine which participants held these views, why, and what experiences might have contributed to them. Many participants held two common misconceptions before participation in the workshop, but they reported changes in their understanding after participating in the workshop. Before starting the workshop, the survey revealed that 70% (14 out of 20) of the educators agreed with the statement, “Care should be taken not to give students with disabilities unrealistic goal expectations which will inevitably result in frustration when they try to find employment,” while only 45% (9 out of 20) left with that idea. Furthermore, 50% of workshop participants agreed with the statement, “Care must be taken not to frustrate students with physical disabilities in science because they are more likely to become frustrated and give up,” prior to the workshop (slightly higher than 34% reported by Norman et al., 1998). However, only 1 participant held that belief by the end of the CLASS experience.

REFERENCES Bargerhuff, M.E., & Wheatly, M. (2004). Teaching with CLASS: Creating laboratory access for science students. Teacher Education and Special Education, 27(3), 318-321 Bargerhuff, M.E., Wheatly, M., & Kirch, S.A. (in press). Collaborating with CLASS: Creating laboratory access for science students with disabilities. Electronic Journal of Science Education. Bogdan, R., & Bilken, S.K . (2003). Qualitative research for education: an introduction to theory and methods. Boston: Allyn and Bacon. Caseau, D., & Norman, K. (1997). Special education teachers use science-technology-society (STS) themes to teach science to students with learning disabilities. Journal of Science Teacher Education, 8(1), 55-68. Cawley, J., Hayden, S., Cade, E., & Baker-Kroczynski, S. (2002). Including students with disabilities into the general education science classroom. Exceptional Children, 68(4), 423-435. Cochran-Smith, M., & Lytle, S. (1993). Inside outside: Teacher research and knowledge. New York: Teachers College Press. Finson, K.D., Ormsbee, C.K., Jensen, M., & Powers, D.T. (1997). Science in the mainstream: Retooling science activities. Journal of Science Teacher Education, 8(3), 219-232. Gartner, A., & Lipsky, D.K. (2002). Inclusion: A service, not a place: A whole school approach. Port Chester, NY: Dude Publishing. Hamre, B., & Oyler, C. (2004). Preparing teachers for inclusive classrooms: Learning from a collaborative School Science and Mathematics 16 inquiry group. Journal of Teacher Education, 55(2), 154-163. Hammrich, P., Price, L., & Slesaransky-Poe, G. (2001). Daughters with disabilities: Breaking downbarriers. Electronic Journal of Science Education [Online serial] 5(4), article 4. Available: http://unr.edu/homepage/crowther/ejse/hammrichetal.html Johnson, L. R. (2000). Inservice training to facilitate inclusion: An outcomes evaluation. Reading and Writing Quarterly, 16, 281-287.

109

First International Conference on Technology-based Learning with Disability Lipsky, D.K., & Gartner, A. (1997). Inclusion and school reform: Transforming America’s classrooms. Baltimore, MD: Paul H. Brookes. Mastropieri, M.A., & Scruggs, T.E., (1994). Textbased vs. activities-oriented science curriculum: Implications for students with disabilities. Remedial and Special Education, 15, 72-85. Mastropieri, M.A., & Scruggs, T.E., (1998). Enhancing school success with mnemonic strategies. Intervention in School and Clinic, 33(4), 201-208. Mastropieri, M.A., Scruggs, T.E., & Bohs, (1994). Mainstreaming an emotionally handicapped student in science. A qualitative investigation. In T.E. Scruggs & M.A. Mastropieri (Eds.), Advances in learning and behavioral disabilities (Vol. 8, pp. 131-146). Greenwich, CT: JAI Press. Moroney, S.A., Finson, K.D., Beaver, J.B., & Jensen, M.M. (2003). Preparing for successful inquiry in inclusive science classrooms. Teaching Exceptionl Children, 36(1), 18-25. National Committee on Science Education Standards and Assessment, National Research Council. (1993). National science education standards: An enhanced sampler. Washington, DC: National Research Council. Norman, K., Caseau, D., & Stefanich, G.P. (1998). Teaching students with disabilities in inclusive science classrooms: Survey results. Science Education, 82, 127-146. Ormsbee, C.K., & Finson, K.D. (2000). Modifying science activities and materials to enhance instruction for students with learning and behavioral problems. Intervention in School and Clinic, 36(1), 10-21. Reiff, H. B., Evans, E. D., & Cass, M. (1991). Special education requirements for general education certification: A national survey of current practices. Remedial and Special Education, 12, 56-60. Salend, S. J. (1998). Using an activities-based approach to teach science to students with disabilities. Intervention in School and Clinic, 34(2), 67-72. Scruggs, T.E., & Mastropieri, M.A. (1994). Successful mainstreaming in elementary science classes: A qualitative investigation of three reputational cases. American Educational Research Journal, 31, 785-811. Scruggs, T.E., & Mastropieri, M.A. (1996). Teacher perceptions of mainstreaming/inclusion, 1958-1995: A research synthesis. Exceptional Children, 63, 59-74. Scruggs, T.E., Mastropieri, M.A., & Boon, R. (1998). Science education for students with disabilities: A review of recent research. Studies in Science 17 Education, 32, 21-44. Stefanich, G. (Ed.). (2001). Teaching in inclusive classrooms, theory and foundations. Washington, DC: National Science Foundation. Stefanich, G.P., Callahan, W., & Johnson, C. (2001). Assistive technologies, safety and accessibility. In G. Stefanich (Ed.), Teaching in inclusive classrooms, theory and foundations (pp. 147-172). Washington, DC: National Science Foundation. Stefanich, G.P., Holthaus, P., & Bell, L. (2001). The cascade model for managing students with disabilities in science classrooms. In G. Stefanich (Ed.), Teaching in inclusive classrooms, theory and foundations (pp. 115-146). Washington, DC: National Science Foundation. Stefanich, G. P., & Norman, K. I. (1996). Teaching science to students with disabilities: Experiences and perceptions of classroom teachers and science educators. Washington, DC: Association for the Education of Teachers in Science. Tomlinson, C.A. (1999). The differentiated classroom: responding to the needs of all learners. Alexandria, VA: Association of Supervision and Curriculum Development. U.S.

110

First International Conference on Technology-based Learning with Disability Department of Education. (2001). Twentythird annual report to Congress on the implementation of the Individuals with Disabilities Education Act, Washington, DC: Author. Wood, T. S. (2001). Laboratory manual 2001 workshop: CLASS: Creating laboratory access for science students. Dayton, OH: Wright State University, CLASS Project.

111


Descriptions of STEM Workshops for Middle School Students with Specific Learning Disabilities Jiang Zhe, Julie Zhao, Paul Lam, Dennis Doverspike, and Craig Menzemer The University of Akron Akron, OH

ABSTRACT This paper summarizes research funded through the National Science Foundation in support of Science, Technology, Engineering and Math (STEM) workshops for middle school students with specific learning disabilities and typical middle school students. The three major objectives of this pilot project are: 1) encourage students with special learning disabilities and typical middle school students to explore STEM as a future career choice by building their confidence and efficacy in STEM; 2) develop empathy and better appreciation of diversity among students who traditionally enter engineering programs; and 3) develop an understanding and appreciation of diversity, and an elaborated sense of teaching and learning styles, among participants. We accomplished this utilizing exciting hands-on activities designed to spark and capture the interests of participants in the STEM fields. The activities were designed based on the Society of Automotive Engineers’ “A World in Motion”, the use of smart balloons, civil engineering materials, and Information and Communication Technology (ICT). The materials presented at the workshops illustrate aspects of inclusive technology and engineering classroom education intended to help the students succeed thereby building self-efficacy and career interest. Data obtained from the participants and their parents via various surveys were used in the analysis. INTRODUCTION Transitioning from middle school to high school is a critical juncture for any student as they begin to make their own choices regarding preferred academic curricula. These decisions are compounded for middle school students with special learning disabilities, who may also be struggling with the ongoing development of their self confidence and self efficacy, social behavior and academic skills. According to the Ohio Department of Education census of 20042005, the percentage of middle school students with Specific Learning Disabilities (SLD) enrolled in Summit and Stark county public and private schools was 6.3% and 6.9% respectively (a much larger percentage of middle school students were on some type of Independent Education Program, IEP). Of current under-graduate students at The University of Akron, approximately four percent, or approximately 1000, were classified as students with SLD (University of Akron, 2006). However, a relatively small percentage of students with SLD consider or enter the STEM fields. In order to address the low enrollment and subsequent achievement of students with SLD, it is necessary to promote attitudes and strategies for inclusive education in science and engineering. In response to this concern and with funding 112

First International Conference on Technology-based Learning with Disability from the National Science Foundation, The University of Akron, in partnership with the Summit and Stark counties public school systems, developed a program that provided opportunities for SLD and typical middle school students to begin to pursue careers in STEM fields. According to a comprehensive literature review of 66 reports involving science education for students with disabilities (Lent et al, 1994; Mastropieri and Scruggs, 1992), knowledge and learning are facilitated by providing activities-oriented science curricula. Thus, the purpose of this project is to explore whether the use of hands-on educational activities will lead to better learning for students with SLD. In addition to improved learning, such activities should lead to increased self confidence, self-efficacy, and career interest in STEM areas. The hypotheses to be tested were: Hypothesis 1: The hands-on program would increase self efficacy in STEM areas among students with SLD and typical students. Hypothesis 2: The hands-on program would increase self confidence in STEM areas among students with SLD and typical students. Hypothesis 3: The hands-on program would increase career interest in STEM areas among students with SLD and typical students. Hypothesis 4: The inclusion of SLD and typical students in the same program would lead to a greater appreciation of diversity in learning styles among SLD and typical middle school students, college students, and science/special education teachers. Specifically, this paper describes only the summer workshop program that was completed in August of 2006 at The University of Akron. This was part of a larger year long program. DESCRIPTION OF THE STEM SUMMER WORKSHOP Using the Society of Automotive Engineers (SAE) ‘A World in Motion (AWIM),’ smart balloon, sensors, information and communication technologies and civil structure materials, the year 1 summer program provided both lectures and hands-on experience in STEM curricula. The uniqueness of our approach was to present the material in a simplified, but unified manner that linked mathematics to the understanding of science and engineering disciplines. It also integrated several engineering disciplines, while emphasizing the inevitable interaction among these subjects in our everyday life. During the first day of the summer workshop, students participated in the AWIM can crusher competition. The students built the can crusher using the concept of levers. They also learned about different types of simple machines using K’Nex sets. On the second day, the summer workshop focused on civil engineering materials and concrete mix design. The students worked in the Department of Civil Engineering Laboratory where they mixed and cured concrete cylinders. During the third and fourth days, the students were exposed to the central concepts and hands-on experiences in state-of the art space surveillance, sensor technologies and Information

113

First International Conference on Technology-based Learning with Disability and Communication Technology (ICT). The workshop centered on assembling and testing of intelligent balloons equipped with motors, sensors and wireless control. The students also participated in pressure and temperature sensors experiments. The last day of the summer workshop, each team tested and broke duplicate concrete cylinders after three days of curing. Additional concrete cylinders were made and left to cure and were to be tested during the academic year workshops. Data was recorded and testing over time and students were able to observe the role of age or extent of hydration, influence of water to cement ratio, additive, aggregate type and grade on concrete strength. After each workshop, the SLD and typical students, teachers and college students filled out attitude and learning surveys. The schedule of the year one summer is summarized in Table 1. Table 1: Summer Workshop Schedule Day 1: A World in Motion Part 1 Part 2

Part 1

Part 2 Part 1

Part 2 Part 1

Introduction to A World in Motion. Can Crusher Competition: Levers. Introduction to Simple Machines. Using the K’Nex set to build simple and complex machines. Day 2: Civil Structures Workshop Civil engineering materials and applications. Introduction to concrete mix design and demonstration mix with slump tests and cylinder preparation. Laboratory- concrete experiments: mix and curing of cylinders. Day 3 and 4: Smart Balloon and Sensors Workshop Basic concepts of flight aerodynamics. Introduction to temperature and pressure sensors and hands-on activities with sensors. Introduction to wireless communication systems used in the smart balloon. Day 5: Civil Structures Workshop Testing of concrete cylinders after 3 day cure.

PARTICIPANTS The participants of this program are middle school students with SLD and typical middle school students from 6th to 8th grade in Summit and Stark counties. The profile of students with SLD and typical students is summarized in Tables 2 and 3, respectively. During year one of the project, we were able to recruit 11 students with SLD and 15 typical students with a lead time of 2.5 months (May-July). In total, 38% were underrepresented students and 42% were female students. The average grade point average for students with SLD was 3.02 compared to 3.53 for typical students. From our observations, most of the students with SLD were very high functioning and they worked well with typical students, especially in the hands-on activities. Due to unanticipated high interest in this program, the PIs decided to have six teams of students and teacher/mentors instead of five teams as originally proposed. Recruitment even outside of the

114

First International Conference on Technology-based Learning with Disability “normal” academic year was not difficult. Overall, there was no difficulty recruiting college student mentors, but recruitment of science/special education teachers with a lead time between May-July was rather difficult. We decided to recruit a pre-service math/science teacher who was a senior from the Department of Education to participate in the program. Also, one of the teacher participants was a parent of a child with SLD as well as a child advocate in the schools; her involvement with the program really aided in understanding the SLD student population, especially for students who were in the autistic spectrum and anxiety SLD. One of the seven college mentors was from the Office of Accessibility. He was a computer science major with a 3.5 grade point average. We believe that he served as a good role model for the students with SLD. The profile of the college student mentors and science/special education teachers are shown in Tables 4 and 5, respectively. It should be noted that the program was designed to train not only the students but also the teachers and mentors. Thus, at the beginning, some of the teachers and mentors did have difficulties involving all of the students in group activities. The college mentor from the Office of Accessibility tended to try to perform the group activities by himself. While we could have intervened to ensure all students were involved, we saw that as a role of the teachers and mentors. We wanted the teachers and mentors to see the benefit of involving all students in the group activities and to learn over time to involve all the students. We believe that we were successful in meeting this goal. Table 2: Profile of middle school students with SLD ID

Ethnicity

Gender Grade

SI1

Asian

Male

8th

2.0

SI2

White/Caucasian

Female

6th

4.0

Dyslexic/Anxiety

SI3

White/Caucasian

Male

8th

3.9

Slightly slow in multistep repeated

SI4

Black

Female

8th

2.0

Math

SI5

White/Caucasian

Male

6th

3.0

Aspergers’ / Autistic Spectrum

SI6

Hispanic/Latino

Male

7th

3.9

Language Arts

SI7

White/Caucasian

Male

8th

3.1

Speech

SI8

White/Caucasian

Female

6th

3.0

Hearing Impaired

SI9

White/Caucasian

Male

6th

3.0

Written expression

SI10

White/Caucasian

Male

8th

2.8

Speech

SI11

White/Caucasian

Female

6th

2.5

Math & Language Arts

115

GPA

IEP Language Arts / Auditory processing

First International Conference on Technology-based Learning with Disability Table 3: Typical middle school students’ profile ID Ethnicity Gender Grade GPA ST1 ST2 ST3 ST4 ST5 ST6 ST7 ST8 ST9 ST10 ST11 ST12 ST13 ST14 ST15

White/Caucasian Black White/Caucasian Black Asian Pacific Black Black White/Caucasian White/Caucasian Black Asian White/Caucasian White/Caucasian White/Caucasian White/Caucasian

Female Male Male Female Male Female Male Female Male Female Male Male Female Female Male

8th 7th 8th 6th 8th 7th 8th 7th 8th 7th 7th 7th 8th 8th 7th

Table 4: Teachers’ profile Years of Teaching Gender Experience

ID

Ethnicity

CT1

White/Caucasian

Female

18

CT2

White/Caucasian

Female

0

CT3

Black

Male

9

CT4

White/Caucasian

Female

30

CT5

White/Caucasian

Female

Parent SLD mentor

CT6

White/Caucasian

Male

28

116

4.0 2.0 4.0 3.0 4.0 2.9 3.2 4.0 2.8 4.0 4.0 3.5 4.0 4.0 3.5

Remarks Learn more about engineering & technology to help guide my students to careers in science. Gain skills and strategies for teaching mathematics in my own classroom in the future. Help teaching mathematics and technology in my own classroom. Work with special needs students to help to give them a clearer understanding of science through hands-on activities. Gain knowledge of the impacts of hands-on activities with students who has SLD. I hope to gain knowledge of how special needs students respond to "hands on" science activities.

First International Conference on Technology-based Learning with Disability Table 5: College mentors’ profile ID CM1 CM2 CM3 CM4 CM5 CM6 CM7

Ethnicity White/Caucasian White/Caucasian White/Caucasian Asian White/Caucasian Asian Asian

Gender Male Male Male Male Male Male Male

Class Level Senior Sophomore/SLD Senior Graduate student Sophomore Junior Junior

Major Civil Engineering Computer Science Mechanical Engineering Mechanical Engineering Civil Engineering Mechanical Engineering Electrical Engineering

ASSESSMENT Monitoring and assessment were carried out on a continual basis throughout the project. Daily pre- and post tests were administered to assess knowledge in the content areas covered by the workshop and attitude surveys were completed by all students. At the end of each workshop project reaction measures were collected. This involved having each student, college mentor and teacher rate the usefulness of specific activities and materials provided and give a narrative describing the strengths and weaknesses of the activity. In addition, each teacher was assessed on their knowledge and attitudes; they also rated knowledge and attitudes of students. This allowed the instructors to learn the effectiveness of each workshop. At the end of each workshop, meetings were held by the instructors with the consultant to discuss and analyze the results in order to make decisions concerning future interventions and implementation. Based on student surveys and observations, the students both learned from and enjoyed the STEM workshops. All workshops received high ratings; the highest rated was consistently the civil engineering concrete activities, probably due to the many hands-on projects. The ratings were lower for smart balloon and simple machines. The responses from teachers and mentors were very positive. On a 5 point scale, with a 5 being the most positive, the means were consistently above 4.00 and many were above 4.50. Thus, responses from teachers and mentors were very positive. The responses from the typical students were also very positive. Again, they were consistently above 4.00. Most responses from students with SLD were also positive, above 3.00, with most being above 4.00. The knowledge test results show that the performance of students with SLD is not worse than the typical students, implying our hands-on activities and team-environments were effective to master the materials covered in the workshop. Based on the surveys, most of the students with SLD favored the program. From the attitude surveys, the responses from the typical students and students with SLD were somewhat similar. Attitude responses were obtained on a 5 point scale. The questions asked of the teachers and mentors included: 1. I believe the students learned a lot in this workshop. 2. I believe this workshop has positively influenced the participating students’ interest in taking more workshops or school classes like this one. 3. I believe this workshop has increased the participating students’ confidence in their ability to do well in high school classes with similar subject matter.

117

First International Conference on Technology-based Learning with Disability 4. I believe students participating in this workshop enjoyed working with a team on real world projects. 5. I believe it would be very beneficial to conduct this workshop in the class/school where I normally teach. The SLD students’ response tended to be a little more negative than the typical students. This was expected because the students with SLD were less engaged and they did not always totally understand the attitude survey questions. In interpreting the results, it should be realized that the sample sizes were small. There were 9 to 11 responses to the surveys distributed to the college student mentors and teachers. For the typical students there were 13 responses and for the SLD students there were 11 responses. The attitude survey indicated the students enjoyed this workshop. To investigate the long term impact of this program, we plan to follow the academic progress of these students during their high school years. Also, we will track the students' majors when they enroll into colleges. At the conclusion of the summer workshop, surveys were sent out to the parents of the participants. The means and standard deviations for the parental survey results are summarized in Table 6. Inspection of Table 6 reveals that the parent responses were very positive. All responses were above the 4.00 (Agree) rating. Based on a one-sample t-test, all the means except for efficacy were significantly above a 4.00 rating. The efficacy rating was not significantly different from a 4.00 population value. For surveys such as this, that is a very positive response. Table 6: Parent Survey Results for Summer Workshop Variable Mean SD N Child Satisfaction 4.70 .47 23 Child Learning 4.48 .51 23 Child Interest 4.48 .67 23 Child Efficacy 4.09 .79 23 Child Team 4.39 .58 23 Parent Satisfaction 4.72 .42 23 Comparing the responses of the parents of typical students and parents of students with SLD, our expectation was that the students with SLD would have similar attitudes or responses to those of typical students. Table 7 reveals that our expectations were confirmed. The responses of the parents of students with SLD were very positive and very similar to those of the typical parents. Even for the efficacy variable for the parents of students with SLD, the mean rated interest was a 4.0 or agrees to the question of whether “this workshop has increased my child’s confidence in their ability to do well in high school classes with similar subject matter.”

118


Table 7: Parent Survey for Summer Workshop IEP N Mean Std. Deviation Child Satisfaction Yes 11 4.64 .50 No 12 4.75 .45 Child Learning Yes 11 4.55 .52 No 12 4.42 .51 Child Interest Yes 11 4.64 .50 No 12 4.33 .78 Child Efficacy Yes 11 4.00 .89 No 12 4.17 .72 Child Team Yes 11 4.36 .67 No 12 4.42 .51 Parent Satisfaction Yes 11 4.68 .46

Significant No No No No No No

DISCUSSIONS AND CONCLUSIONS Based on our initial surveys and observations, the students both learned from and enjoyed the summer workshop. Although all the workshops received high ratings, the highest rated workshop was consistently the civil engineering concrete activities, probably due to all the hands-on projects. The ratings were lower for smart balloon and simple machines. Qualitative data and observations suggest that an area that needs to be worked on in the future is that of encouraging interaction between typical students and their peers with SLD. Some of the students with SLD had communication problems that interfered with interactions. In addition, the students with SLD had a wide range of limitations, which limited the use of any one solution to the problem of encouraging interactions. Some of the students with SLD were very high functioning, so it is important not to single them out as these students were trying hard to mainstream. One solution that will be considered in the future is rotating students through teams to encourage greater interaction. In order to keep the attention of all the students in focus, especially the SLD students, we limited the lecture portion of the workshop to twenty to thirty minutes during the academic year workshops. In addition, the team approach for the workshop activities established a need for the students to learn fundamental concepts and skills in a very cooperative setting. We found these activities helped students with SLD to develop socialization and communication skills with typical students, mentors and teachers. Overall, the workshop was well received by teachers, mentors, parents and both the SLD and typical student groups. ACKNOWLEDGEMENT This project is supported by the NSF grant RDE-FRI (Award # 0622767). We would like to thank The University of Akron College of Engineering, The Ohio Space Grant Consortium and the Summit and Stark County school systems in support of this project.

119

First International Conference on Technology-based Learning with Disability REFERENCES Lent, R. W., Brown, S. D., & Hackett, G. (1994). Toward a Unifying Social Cognitive Theory of Career and Academic Interest, Choice, and Performance. Journal of Vocational Behavior, v 45, p. 79-122. Mastropieri, M. A. & Scruggs, T. E. (1992). Science for Students with Disabilities. Review of Educational Research, v 62, n 4, p. 377-411 University of Akron. (2006). Unpublished Data obtained from the Office of Accessibility.

120


Equity in Accessing Academic Content: Support for General Educators Teaching Students with Disabilities in Inclusive Classrooms Heidi Cowan Ohio State University Dr. Mary Ellen Bargerhuff Wright State University Dr. Jackie Collier Wright State University ABSTRACT The following paper will introduce the reader to the Ohio Resource Center’s Lesson Adaptations for Student Success (ORCLASS), which is a pilot project designed to help give educators the needed tools to include students with disabilities into their general education classroom. To do this ORCLASS makes specific recommendations on how to adapt a given lesson plan. The quality lesson plans used are already well established in the educator community, and are provided on the Ohio Resource Center’s (ORC) website (www.ohiorc.org). ORC has an established community of educators who routinely access the site for examples of best practice. The lesson plans found in the ORC depository are all online resources and are peer-reviewed and aligned with the national and state standards in the respective content areas. ORCLASS makes specific recommendations to help educators include their students who have/are blind/visually impaired, deaf/hard-of-hearing, autism, learning disabilities, orthopedic impairments, and behavioral disabilities. These adaptations apply to 7th-12th grade lessons in science, mathematics, and reading. INTRODUCTION This paper and subsequent presentation focuses on the right of students with disabilities to equitable learning opportunities in the general education classroom. In examining the issue of equal access to a quality education, one segment of our nation’s children that continues to struggle is that of students with disabilities. Persons with disabilities have struggled for decades for equitable treatment in their education. For example, many students with disabilities were routinely excluded from public school until the passage of the Education of All Handicapped Children Act (EHA) in 1975. The EHA, first reauthorized as the Individuals with Disability Education Act (IDEA), and now reauthorized as the individuals with Disabilities Education Improvement Act of 2004 (IDEIA), mandates free and appropriate public education for all students in the least restrictive environment (Kirch, Bargerhuff, Turner, & Wheatly, 2006). While many factors contribute to these inequities, one particular barrier that is a fundamental 121

First International Conference on Technology-based Learning with Disability responsibility of teacher preparation programs and school districts is that of insufficient skill development in educators who are charged with teaching a diverse population. Kirch, Bargerhuff, Turner & Wheatly (2006) report that more than fifty percent of the science and mathematics educators who attended their summer workshop believed they had inadequate training in the area of working with students with disabilities. Ninety percent of participants of these summer workshops reported not having preservice training or experience in working with students with disabilities in a science laboratory (Bargerhuff & Wheatly, 2004). The concept of Least Restrictive Environment has moved the focus of education for those with disabilities from segregated environments to ones of inclusion within the general education classroom. This move demands training and skill development opportunities for general educators. Bargerhuff and Wheatly (2004) note “Formal preparation of science teachers has not kept pace with changing legislation that mandates access of students with disabilities to general education curriculum” (p. 139). For an inclusive education program to be effective educators must have the skills to include students to the fullest extent and they must have access to the resources and guidance needed for such a task. In short, we must equip today’s teacher leaders with the tools necessary to accommodate individual strengths and needs. At Wright State, the CLASS Project (Creating Laboratory Access for Science Students), a collaborative among the College of Education and Human Resources, the College of Science and Mathematics, the Office of Disability Services, and several partner schools, has been addressing the issue of equitable learning opportunities for students with disabilities since 1998. This program was funded by the National Science Foundation and directed by Michele Wheatly, current COSM Dean. The project has focused on providing needed resources and skill development opportunities for science and mathematics educators who work with students with physical and learning disabilities. The projects promoted by the CLASS Project have focues primarily on professional development opportunities including a summer workshop and miniworkshops for individual school districts. Mini-grants were also awarded to school teams who presented a focused plan of increasing inclusion opportunities for their students within the sciences (Bargerhuff & Wheatly, 2004; Lunsford & Bargerhuff, ; Kirch, Bargerhuff, Turner, & Wheatly, 2005; Kirch, Bargerhuff, Cowan, & Wheatly 2006). An unmet CLASS priority was that of curriculum development. Ohio Resource Center’s Lesson Adaptation for Student Success (ORCLASS) was initiated when the staff members from CLASS saw the need for educators to have guidance in how to adapt lessons so as to allow all students full access to the content, Since CLASS did not have the mechanism needed to branch out into this area of education they reached out to the Ohio Resource Center to form a partnership. The development and early stages of this program are the focus of this paper and presentation. PROGRAM DESCRIPTION ORCLASS is a program that strives to give educators easy access to lesson plans that are suitable or easily adapted for students with disabilities. Our goal is for educators to feel comfortable using the lesson along with the adaptations provided so that including their students with disabilities becomes not only possible, but feasible. We understand that educators’ have hectic schedules and are not always able to research best practices related to a particular group of students. Therefore, ORCLASS works to provide the recommendations in a way that complements what is already being done within the classroom. To provide such a service we teamed with the Ohio Resource Center (ORC), a virtual resource center that identifies, evaluates,

122

First International Conference on Technology-based Learning with Disability and disseminates peer-reviewed lesson plans in math, science, and reading. These lessons are aligned with national and state academic content standards and are peer-reviewed. Lessons that meet rigorous standards are posted on the ORC website and are free to anyone who chooses to access them. ORCLASS directs teams of teacher leaders who have expertise in the content areas and in special education to make a new layer to the lesson plans already accessible on the ORC website. These educator teams review lessons and suggest adaptations for students with learning disabilities, autism, blindness, deafness, orthopedic impairments, and emotional disturbances. Suggestions are given for each disability category separately and range from general adaptations to those that are specific to the given lesson. It is our hope that educators who use this service will be able to use the lesson adaptations as a model and will ultimately be able to make similar adaptations to other lesson plans they are using within the classroom. The idea of ORCLASS was started by an informal conversation between a CLASS staff member and a representative from the ORC. A grant from the National Science Foundation was secured for the adaptation of the science and mathematics lesson plans found on the ORC website and additional funding through a program entitled AdLit was secured for adapting the literacy lesson plans. This program is just completing a pilot stage complete with successes and challenges common to new endeavors. Because of the newness of the program we decided to limit initial adaptations to lesson plans geared toward middle and high school classes. These lessons were chosen because the CLASS Project has focused on this age range in the past. The arrangement of this program called for the setup of pairs or teams of educators which included at least one content specialist and at least one special educator. The project began with an initial hiring process that netted twelve educators. These employees were called “Abstractors for Accessibility” and it was their duty to work in teams to write up recommended adaptations for given lesson plans. The mix between general educators and special educators and the mix between science, mathematics, and reading educators were not as even as we had hoped for, but some employees had experience in both a general education area as well as in special education. This helped to ensure there was appropriate experience within each team. TRAINING Training for the abstractors included a four-hour workshop where the program was introduced and teams were decided. Exercises helped orient the employees to the ORC website and the lessons they would be working with. We also had an exercise which included reviewing a lesson and making adaptations so that the abstractors could get to know their teammates and would have a reference as to what types of adaptations were available for the sample lesson. Focused training on what adaptations and equipment/tools were available for educating various students with disabilities was given in the following manner. Abstractors were paired up and were asked to compile a master list of accommodations and best practices used for one particular group of students (i.e. for students who are blind or visually impaired, for students who are deaf or hard of hearing). Teams were also asked to research a particular piece of equipment or software package that is commonly used by persons with disabilities. Such items included speech recognition software, screen enlarging software, accessible microscopes, closed circuit televisions, etc. These teams were given two weeks to compile this information and then a large group meeting was held where each pair of employees reported on their findings and provided their colleagues with a print copy of their work, for future reference. This exercise allowed each abstractor to delve deeply into at least on area of disability, but also allowed them to have a wide

123

First International Conference on Technology-based Learning with Disability variety of ideas for recommending adaptations for students who were in other disability categories. This was important because each educator team would be responsible for making adaptations for all six-disability categories for their lessons. Therefore it was important for each abstractor to be familiar with best practices for each group of students. INITIAL DIFFICULTIES A couple of things slowed the progress of the project. First, the timing of hiring and training was, unfortunately, near the end of summer and there was not much time before educators were back to having fulltime teaching responsibilities. This position was originally meant to be a summer position for educators, but due to some delays in the hiring process, the educators didn’t really get started until the beginning of August, which was close to the start of the school year. This severely limited the time some of the abstractors had to devote to the task. Secondly, since this was a pilot project, there was no set format for adapting the lessons. The intent was to have the employees determine the best design for the adaptations and to make models of what an adapted lesson should look like. In retrospect, it would have been more productive if a format was decided upon by the ORCLASS staff and then modified once abstractors were able to use the process and find out the strengths and weaknesses. How it turned out was confusing to the abstractors and it seemed as though no one was on the same page. We tried to remedy this by setting a standard as to what to include in a report of a lesson’s adaptation, but this proved to be too general. We focused on how the information was to be presented, but did not stress to what level the information was to be conveyed. This led to confusion as to how specific recommendations should be and to what extent they should be described. Overall, there was a lack of clear guidance which was discouraging to some of the abstractors. This led to two abstractors dropping out of the program and others becoming less involved in the program. As are result the timing of the project was significantly delayed. The pilot project was intended to be one-year in length with a completion goal of approximately 150 lesson plans, in each subject area. In December of 2005 we asked for a oneyear extension. We did not have nearly as many lesson plans as we had anticipated and we devised an action plan to address this challenge. Fourteen new abstractors were hired at the end of May 2006. This allowed for training to begin at the beginning of June and that allowed for the abstractors to move directly from their teaching responsibilities to ORCLASS for the summer months. This, and the introduction of an online reporting system, discussed next, significantly increased the efficiency by which lesson adaptations were made and reported. BEHIND THE SCENES As the educator teams were being established and trained and lesson plans were being reviewed and adaptations written, ORCLASS staff members were working alongside the staff of the ORC to discuss how the additional information would be incorporated into the existing site and what information would be important to include for educators to get the most out of the information provided. There were also extensive meetings between a CLASS staff member and the computer programmers of the ORC to design a template to be used to make the adaptations and for the final look of the adaptations, once they were posted to the live part of the website. This process took much longer than anticipated, but once this system was up and running, it made the input of adaptations much smoother for the abstractors. The introduction of this online

124

First International Conference on Technology-based Learning with Disability system proved to be invaluable to the standardization of the adaptations for the lessons. After this was introduced, the abstractors were able to finish their recommendations to a lesson much more efficiently. This system allows for an abstractor to place a check mark by general recommendations that are appropriate for the given lesson plan. Along with check marking appropriate adaptations, abstractors were able to type in additional suggestions that would specifically relate to the lesson at hand. This allowed for maximum versatility while at the same time having an easy way to identify general adaptations that are common from one lesson to another. CONCLUSION The lesson adaptations we have discussed within this paper are not yet available to the public. Our hope is to have them available by the end of summer 2007. To date we have approximately 370 adapted lesson plans that are going through a final editing process. Once this process is complete we will be able to present our work on the ORC website. The development of this project has been educational for our staff. In hindsight we can see where the shortcomings of the program were and so we will be better positioned to not only complete this project, but to expand our efforts in the future. The effort is one that is worthwhile and will serve educators well by equipping them include all of their students into classroom activities. FUTURE PLANS Once these lessons are accessible to the public and have been available for several months, we will conduct formal and informal studies on the usefulness and appropriateness of the adaptations. This will be done through teacher surveys and focus groups along with classroom observations. We will also expand our work to lower grade levels and eventually make adaptation recommendations for all the lessons found on the ORC website. The ORC is open to anyone who has access to the internet; however, the lessons are aligned with the Ohio State Standards and the website is mainly advertised to educators within Ohio. We would like to see our work reach a broader audience. Therefore, if research shows a usefulness of this service, we will partner with other organizations that provide educators with classroom resources and to do similar work with those associations.

125

First International Conference on Technology-based Learning with Disability REFERENCES Bargerhuff, M. E., & Wheatly, M. G. (2004). Teaching with CLASS: Creating Laboratory Access for Science Students with Disabilities. Teacher Education and Special Education, 27(3), 137-140. Kirch, S., Bargerhuff, M. E., Turner, H., & Wheatly, M. (2005). Inclusive science education: classroom teacher and science educator experiences in CLASS workshops. School Science and Mathematics, 105(4), 175-196. Kirch, S., Bargerhuff, M. E., Cowan, H., & Wheatly, M. (in press). Reflections of educators in pursuit of inclusive science classrooms. Journal of Science Teacher Education. Lunsford, S., & Bargerhuff, M. E. (2006). A project to make the laboratory more accessible to students with disabilities. Journal of Chemical Education, 83(3), 407-409.

126


Accessibility of Higher Education Web pages in the Northwestern U.S.: Current Status and Response to Third Party Outreach Terry Thompson Sheryl Burgstahler, Ph. D. Elizabeth J. Moore, Ph.D. University of Washington ABSTRACT In a study of web page accessibility to people with disabilities, colleges and universities in the northwestern U.S. were assigned to one of three groups ("extensive" outreach; "moderate" outreach; and no outreach) and were offered varying degrees of training and/or support with web accessibility depending on their group assignment. At each institution, key web pages were measured for accessibility three times in six months. Baseline assessments on 14 accessibility checkpoints reveal that accessibility is highly dependent on the specific measure of accessibility. Results range from nearly all measured institutions being fully accessible on some checkpoints (avoidance of flickering content) to nearly all measured institutions being fully inaccessible on others (e.g., titles on frames). Accessibility ratings differ by Carnegie classification with the accessibility of web pages in smaller, specialized schools most in need of improvement. Followup assessment showed that over six months, accessibility improved on basic, easy-to-implement accessibility checkpoints, but worsened on checkpoints associated with emerging, dynamic web applications. Training and support was not sufficient to counteract this negative trend. Results also suggest that a simple, concrete message with specific instructions for implementation can be a successful intervention strategy for improving web accessibility. INTRODUCTION The Internet plays an integral role in an increasing number of functions in the development, management, and delivery of academic courses, student services, and administrative functions within postsecondary education institutions. However, many web pages are designed in such a way that they are inaccessible to people with disabilities. The inaccessibility of campus web pages is especially significant as increasing numbers of people with disabilities enroll in postsecondary institutions (Henderson, 2001; National Council on Disability, 2000). Numerous published studies have reported that web pages at U.S. postsecondary institutions tend to incorporate features that are inaccessible to people with disabilities (e.g., Hackett & Parmanto, 2005; Schmetzke, 2003; Thompson, Burgstahler, & Comden, 2003; Williamson, 2003; Zaphiris, & Ellis, 2001). A variety of resources are available for learning about web accessibility, including conferences, websites, online courses, and online discussion groups. However, web developers must proactively locate and make use of these resources. Existing literature and the experiences of the authors of this article suggest that many higher education web developers do not take this initiative, either because they are unaware of web accessibility issues or because they have other issues competing for their time. For these web developers to improve the accessibility of their

127

First International Conference on Technology-based Learning with Disability web resources, an internal stimulus, such as a mandate from higher administration, or an external stimulus, such as an outside entity reaching out to them offering accessibility suggestions and support, may be required. Researchers in the study reported in this article explored the effect of external intervention on change in the accessibility of postsecondary websites. Research Questions The current study addresses five research questions. The first three relate to the current accessibility of targeted websites. The last two relate to the effectiveness of third-party outreach. 1. How accessible overall and with respect to specific accessibility issues are the key web pages of postsecondary institutions in the Northwestern United States? 2. Is there a significant difference in accessibility of web pages across Carnegie classifications? 3. Is there a significant difference in accessibility across key web pages (e.g., are institutional home pages significantly more or less accessible than departmental pages)? 4. Is the accessibility of higher education web pages improve when outreach is provided from a third-party offering free web accessibility suggestions, resources, and ongoing communication? 5. Are certain types of outreach more effective than others in bringing about measurable improvement in web accessibility? METHODS Participants The Northwest Alliance for Access to Science, Technology, Engineering and Mathematics, or AccessSTEM, hosted at the University of Washington in Seattle, serves four states in the Northwest – Alaska, Idaho, Oregon, and Washington. AccessSTEM supported the current study with funds from the National Science Foundation. The population for the study was 127 higher education institutions from the AccessSTEM region. Participating institutions were among the 133 schools recognized by the Carnegie Foundation, across all Carnegie classification categories. (Carnegie Foundation, n.d.). The population includes 57 associate's colleges, 14 baccalaureate colleges, 26 masters colleges and universities, 7 doctoral/research universities, and 23 institutions that fall into other categories, including one tribal college. Six additional institutions were excluded from the study because their inclusion could potentially introduce bias into the sample characteristics: the researchers' own institution (University of Washington) was excluded because the researchers and others have been involved in promoting web accessibility internally for over a decade; the five Northwest campuses of ITT Technical Institute were excluded because their web pages are homogenous and are maintained by a central national entity. A set of "critical" web pages was selected for study from each institution, consistent with previous research by Thompson, Burgstahler, and Comden (2003). The term "pages" is used rather than "sites" in this paper because the unit of measure most commonly assessed was a single page. However, in cases where the functionality of a page is dependent on access to multiple pages, each required page was included in the assessment. For example, an assessment of the "search page" actually includes an assessment of both search and results pages, yielding one score for the accessibility of the "search" function. Pages deemed “critical” (primarily based on traffic volume across several institutions) include the institution home page; search page; page listing departments and units; campus directory of faculty, staff and/or students; admissions home page; course listings; academic calendar; employment home page; job listings; and library home page. Because of the interests of the National Science Foundation, four additional pages

128

First International Conference on Technology-based Learning with Disability representing science, technology, engineering, and/or mathematics (STEM) were selected when available. Typically these pages were the home pages for academic colleges or departments representing these disciplines. At some institutions, no web page could be found that matched a particular category. In other cases, the web page in a particular category required user authentication and was therefore excluded from the study. Procedures To measure web accessibility, the researchers developed a set of 14 web accessibility checkpoints, based on Priority 1 checkpoints from the Web Content Accessibility Guidelines 1.0 (World Wide Consortium, 1999) and the Electronic and Information Technology Accessibility Standards (Office of the Federal Register, 2000). Checkpoints were designed to reflect accessibility challenges that the researchers commonly observe on higher education web pages, and to be easy to test, even by testers who do not have extensive technical knowledge. Each web page was assigned a rating for each checkpoint using the following 3-point scale: 1 = Site has failed to implement this checkpoint. 2 = Site has partially implemented this checkpoint. 3 = Site has fully implemented this checkpoint. Following is a list of checkpoints used in the study. Supporting documentation was developed for researchers and accessibility testers to use during assessments, providing specific instructions on how the 3-point scoring system should apply to each checkpoint and including examples. A modified version of the documentation is available at http://staff.washington.edu/tft/rubric.php. 1. Are frames appropriately titled? 2. Is alternate text for images present and sufficiently equivalent to the graphic content? 3. Are form elements explicitly associated with labels? 4. Is information in PDF available in other more accessible formats as well? 5. Are all links and navigational elements present and contextually appropriate by keyboard? 6. Does the site avoid conveying meaning with color alone? 7. Are data tables marked up as required? 8. Is multimedia content captioned (or if audio only, transcribed)? 9. Is flickering content avoided? 10. Is a skip navigation link present if needed? 11. Is the page functional when scripts are disabled? 12. Is the page functional when style sheets are disabled? 13. Does link text provide a reasonable description of the link target? 14. If a page requires a timed response, can users request more time? A team of five web accessibility testers was assembled. The team included graduate students, recent college graduates, and a former corporate web developer. All team members had experience developing web pages, and were comfortable to proficient with Hypertext Markup Language (HTML). None had extensive web accessibility knowledge when they initially joined the team. All team members received training, which included mock evaluations. Trainees began conducting evaluations independently only after their results were considered by the researchers to be consistently reliable when compared to results of experienced evaluators. Three sets of web accessibility evaluations took place at approximately three-month intervals during the period from December 2004 to December 2005. The first was a benchmark assessment which took place prior to making contact with institutions. During this assessment, 1576 web pages from all 127 institutions were evaluated in order to measure the overall state of

129

First International Conference on Technology-based Learning with Disability accessibility of critical higher education web pages in the Northwest. This was the only evaluation that included all of the available key web pages for each institution. The three-month and six-month follow-up assessments included only home pages and were designed to measure changes in accessibility over time. To measure the effect of third-party outreach and support, researchers at the University of Washington worked in collaboration with the Region 10 ADA & IT Center, which was funded by the National Institute for Disability and Rehabilitation Research at the U.S. Department of Education to provide web accessibility outreach and support specifically to Associate's and Tribal Colleges. This study initially divided institutions into three groups: two experimental groups, whose members were contacted by researchers, and a control group, whose members were not contacted by researchers. Letters of outreach were sent via email to members of both experimental groups following the first and second assessments. The first letter explained the research project, offered general but tailored suggestions regarding web accessibility, listed available resources including an online discussion list specifically set up for this project, and offered further individualized assistance. Outreach to the two experimental groups differed only in the level of further assistance that was offered. One group (the "moderate offer" group) was offered an email address to contact with questions about accessibility, while the other group (the "extensive offer" group) was offered oncampus training and consultation from the Region 10 ADA & IT Center. The second letter to members of both experimental groups was sent after all institutions had been evaluated for a second time, and simply served as a reminder to recipients about the project and available resources, including individualized support. The second letter, like the first, differed only in the level of support that was offered. The third letter was sent only to members of the "moderate offer" group. It was sent in mid-to-late August, just prior to the start of a new academic year for many institutions. The letter was very succinct, and encouraged recipients to "take a moment this summer to implement at least one quick accessibility solution." The letter then provided a very specific recommended solution, tailored to each institution, including steps for implementation. RESULTS By the end of the study, 19 institutions contacted the researchers regarding web accessibility. Seven institutions sent polite emails expressing thanks and/or intent to address accessibility problems, 6 institutions sought and received moderate levels of assistance (e.g., discussed specific web accessibility issues via email), and 6 institutions received extensive training and support by teleconference or in person. The institutions that chose to contact the researchers spanned multiple assigned outreach groups. Therefore, this paper reports on the effect of actual support received, rather than on the effect of assignment to a particular outreach group. In the following paragraphs, results are organized by research question. Research Question 1. How accessible are the key web pages of postsecondary institutions in the Northwestern United States overall and with respect to specific accessibility issues? Analysis of checkpoint ratings reveals few significant differences across types of pages, so different types of pages (e.g., home pages, search pages) were analyzed together. However, checkpoints were not combined into an "overall accessibility" score because the relative importance of each accessibility checkpoint is highly dependent on context. For example, images with missing alternate text may present a barrier to a blind user on one web page, but may be of little consequence to that same user on another page.

130

First International Conference on Technology-based Learning with Disability Figure 1 shows the average rating for each checkpoint. The only checkpoint on which almost all pages show full accessibility is the checkpoint measuring avoidance of using flickering content. Other checkpoints on which the mean rating across all web pages was 2.0 or higher are: content accessible with CSS disabled (2.8), all features accessible using keyboard (2.7), information not depicted using color alone (2.7), content accessible with scripts disabled (2.7), meaningful link text (2.6), and alternate text for images (2.0). Institutions performed relatively poorly on the remaining checkpoints, including user control over timed response (1.6), captions or transcripts on multimedia (1.4), accessible markup on forms (1.2), accessible markup on tables (1.2), a skip navigation link on pages with redundant navigation (1.2), accessible alternatives to PDF (1.1), and titles on frames (1.0).

Research Question 2. Is there a significant difference in accessibility of web pages across Carnegie classifications? Yes. At the baseline assessment significant differences among at least two Carnegie classifications were observed on nine of the fourteen checkpoints (Table 1). Doctoral and masters institutions more frequently have the higher average ratings on these checkpoints. This is especially true for institutions classified as "doctoral/research universities-extensive", which have the highest mean rating on five checkpoints. In contrast, smaller specialized institutions tend to have low mean ratings across multiple checkpoints. Table 1. Checkpoints for which significant differences across Carnegie classifications were observed. 131

First International Conference on Technology-based Learning with Disability Checkpoint

DF

F ratio 9.6***

Results of post hoc analyses High Mean Rating Low Mean Rating 1. Titles on Frames 7, 69 Doctoral-Ext (2.0) All other Carnegie classifications (1.0) 2. Alternate text for 12, 1481 8.2*** MA-I; Doctoral (2.2 to BA/General; Art, music, images 2.3) design; Theological; Engr/Tech (1.6 to 1.8) 3. Accessible markup 12, 756 5.9*** Doctoral/Ext, Health BA; MA-II; Med school; on forms (1,7, 1.8) Art, music, design; Engr/Tech; Theol; Tribal, Doctoral/Int (1 to 1.1) 5. Keyboard 12, 1507 5.7*** Associate’s; BA-Liberal Theological; Medical accessible features Arts; MA-II; Doctoral; school (2.3, 2.5) Tribal (2.8 to 3.0) 9. Flickering content 12, 1549 2.5** Most pages received a 3 Art, music, design; avoided Doctoral-Ext (2.96); MA-II (2.95) 10. “Skip navigation” 12, 1370 6.6*** Art, music, design; BA; MA; Doctoral-Int; link provided if Health (1.5, 1.7) Med schools; Engr/Tech; needed and Tribal (1.0 to 1.05) 11. Content accessible 12, 1549 7.4*** MA; Doctoral; BA; Med school, Engr/Tech; w/ scripts disabled Associate’s; Tribal; Theological; Art, music, Health (2.6 to 3) design (1.9 to 2.5) 12. Content accessible 12, 1229 5.8*** BA; MA-II; Health; Engr/Tech (2.6) with CSS disabled Tribal (3.0) 13. Meaningful link 12, 1500 6.5*** MA; Doctoral; Theological; BA; Health; text Engr/Tech; Tribal (2.7 to Art, music, design (2.2 to 3.0) 2.6). Research Question 3. Is there a significant difference in accessibility across key web pages (e.g., are institutional home pages significantly more or less accessible than departmental pages)? The only difference found that depended on type of page for any of the checkpoints was the checkpoint related to alternate text for images. On this checkpoint, the library home page had a mean rating of 2.2; the institution's home page, search page, and academic programs page all had mean ratings of 2.1; the departmental science and technology pages each had mean ratings of 1.9; and the departmental mathematics page had a mean rating of 1.8. All other pages had mean ratings of 2.0. The difference between the high end (the library home page) and the low end (mathematics page) is statistically significant (F(13, 1480)=1.8; p

LWD - 07 - CiteSeerX

LWD - 07 - CiteSeerX

Suggest Documents

07 - CiteSeerX

07 - CiteSeerX

07-A Review - CiteSeerX

07 wawrzynska xp - CiteSeerX

07 2007 - CiteSeerX

07-A Review - CiteSeerX

07 - Namiotko - CiteSeerX

07 ranganathan.pmd - CiteSeerX

07 Thonse - CiteSeerX

07 Apostolou - CiteSeerX

07. Capitulo 8 - CiteSeerX

07 tashkin / review - CiteSeerX

Precision Agriculture '07 - CiteSeerX

07 SENINI documento - CiteSeerX

07 d onghia - CiteSeerX

07.Broad (1) - CiteSeerX

07 wolpoff & lee - CiteSeerX

07-Faixova.vp:CorelVentura 7.0 - CiteSeerX

07 ranganathan.pmd - CiteSeerX

07 PRB346526_rev1.indd - CiteSeerX

Splitting hairs 07 - CiteSeerX

07 kayworth.pmd - CiteSeerX

07 Manasse - Stanca_279_308 - CiteSeerX