Available online at www.sciencedirect.com
Procedia Technology 5 (2012) 839 – 848
CENTERIS 2012 - Conference on ENTERprise Information Systems / HCIST 2012 - International Conference on Health and Social Care Information Systems and Technologies
A Usability Testing Experiment For A Localized Weight Loss Mobile Application Selma Limam Mansar*, Shashank Jariwala, Maahd Shahzad, Aysha Anggraini, Nawal Behih, Amna AlZeyara Carnegie Mellon University Qatar, Doha, PO Box 24866, Qatar
Abstract This paper introduces a context-appropriate mobile application for sustainable weight loss. Overweight and obesity are acknowledged to have become a worldwide health matter. Addressing weight loss and sustaining efforts remains in many ways a fragile undertaking. Strategies will vary by age group, gender, and social context. Moreover, the cultural, traditional ecosystem will impact weight loss strategies. In this paper, we discuss contributions in the literature for technology-based weight loss support. We design a mobile application that leverages three strategies from proven behaviour change theories (increasing awareness of the aims of dieting, fostering motivation and self-efficacy, and impacting dieters’ attitudes). We adapt the application to the local context of a middle easterner’s society by conducting a usability testing experiment with potential users of the application. We also apply principles of localization to derive an appropriate application. Beyond the applied usage of the application, the paper contributes to the currently scarce body of literature on Arabic based mobile development.
© 2012 Published Elsevier Ltd. SelectionLtd. and/or peer review under responsibility of CENTERIS/SCIKA © 2012 by Published by Elsevier Selection and/or peer-review under responsibility of Association for Promotion and Dissemination of Scientific Knowledge CENTERIS/HCIST.
Keywords: Weight management; healthy life style; mobile application; behavior change theories; mobile phones usability testing; mobile phones localization; Middle East
* Corresponding author. Tel.: +00-974-4454-8650; fax: +00-974-4454-8410. E-mail address:
[email protected].
2212-0173 © 2012 Published by Elsevier Ltd. Selection and/or peer review under responsibility of CENTERIS/SCIKA - Association for Promotion and Dissemination of Scientific Knowledge doi:10.1016/j.protcy.2012.09.093
840
Selma Limam Mansar et al. / Procedia Technology 5 (2012) 839 – 848
1. Introduction In this paper we propose a mobile application that addresses sustainable weight loss [1] and is culturally and contextually adapted to a Middle Eastern audience. The application implements three levels of support for changing dieters’ knowledge, attitude, and motivation to achieve a healthy weight and a healthy life-style. The three levels of support are as follows: (1) Strategy One, “Increase awareness of dieting aims”: We will remind dieters of their set dieting goals by sending them tailored text messages. Based on Strobe [2] and Strobe et al.’s [3] theory, we expect that text messages will result in improved dieting behavior and indirectly result in weight loss, but not necessarily in changed eating attitudes or body image. (2) Strategy Two, “foster motivation and self-efficacy”: We will provide support with sub-goals and visualizing progress by displaying sub-goals and current state of achievement. We expect visual sub-goals tracking will contribute to changing motivation to achieve the final goal and indirectly contribute to weight loss, but not necessarily foster changed eating attitudes or body image. (3) Strategy Three, “Impact dieters’ attitudes”: We will provide means for social support through technology. One aspect of that support are the tailored text messages from the dietician implemented for strategy one. As another aspect, we incorporate social networks (group discussion boards, chatting) and protect access to appropriate participants (dieters with similar goals and caregivers). We hypothesize it will contribute to changing motivation to achieve the final goal. The theoretical background for the recommended level of support was presented in [4] and draws from theories on behavior change. In the sequel, we start by discussing the current state of the art on mobile technologies and their use in weight loss and in promoting a healthy life style; we then describe a fullydeveloped mobile application that exhibits the three aforementioned supporting strategies. The remaining sections of the paper present how the application was validated (section 4) and modified (section 5) to fit a culturally specific audience: The paper enhances the state of the art on both mobile application with a weight loss perspective and on directions for the localization of mobile applications. 2. Mobile Technology Background Mobile technologies have been demonstrated to be effective for supporting dieting behavior. For example, there is evidence that tailored health information delivered via mobile technology or via web can be effective for improving health behavior and/or reducing weight. In one study, tailored health information for online supermarket service impacted shoppers to purchase healthier alternatives even after the service was terminated [5]. Food labeling in a college dining hall supported students’ healthy food choices in another study [6]. Internet/video-delivered 40-min sessions on diet and activity improved students’ health behavior compared to control [7]. More specifically, tailored text-messages have been demonstrated to support dieters’ weight loss efforts either compared with control or with other methods of delivery. A meta-study by Webb and colleagues [8] found that sending dieters SMS (Short Message Services) significantly enhanced web-based interventions on weight loss. Patrick and colleagues [9] found that SMS and MMS (Multi-Media Messaging Services) were the most effective among tailored health interventions for obese people in terms of weight loss. Similar effects are reported for weekly tailored emails from experts [10], bi-weekly SMS exchange [11], Palmtop computerbased Ecological Momentary Interventions (EMI) [12], and sending weekly emails on exercise and eating diaries [13]. Other studies indicate that focusing on specific sub-goals supported by mobile technology can foster dieters’ weight loss: Transmitting blood pressure, weight, and responses to goal achievement questions via phone resulted in marginally significant BMI changes compared to control [14]. Weekly emails on a patient’s
Selma Limam Mansar et al. / Procedia Technology 5 (2012) 839 – 848
stage and specified goals improved eating habits [15]. Finally, text messages sent to adolescents as part of a diet plan are well accepted [16]. Mobile technology has the potential to impact dieters’ health behaviour by using social influence, or more specifically, mechanisms of social support from the nutritionist’s side or social referencing from the dieter’s side. SMS from experts being more effective for weight loss than a commercial website [10]. This might be at least partially explained by social impact. Other studies show that social referencing impacted overweight children’s eating and exercise patterns [17][18]. Unfortunately, current studies on use of mobile technologies do not yet allow estimating the impact of implemented components in isolation because of methodical issues. Therefore, in our current study, we introduce one component at a time and measure potential effects after a period of time before introducing the next component. Indeed, most of these studies did not systematically investigate how weight loss was achieved using behaviour change interventions; content and method of delivery were confounded; or other methodical issues make it hard to evaluate potential benefits of mobile technologies beyond testifying that they might be beneficial in addition to delivered cognitive behavioural therapy. Woolford et al. [16] present an interesting approach to the problem of obesity with adolescents by providing a library of text messages and images sent through phone, with a tailoring and an automated delivery method. While indicating enthusiastic response from adolescents on receiving the messages, the study did not provide any insights on measurement of impact on weight loss or even behaviour change. In Lee at al. [19], the research describes a diet planner and a diet game. While the application looks attractive, there was no theory driving the study, the participants’ pool was less than 20 for each group, with a significant variance in age groups. The study was also conducted for six weeks only (not enough to ascertain change). Mattila et al. [13] describe the usage of a mobile phone to record self-observation as a mean for self-motivation with diet and suggest it is a valid tool for observing eating habits and behaviours. However, the study design is again questionable (clear gender imbalance, wide BMI range including overweight and obese participants, no control group). In the Kornman et al. [20] study, the experiment, which led to mitigated results on adoption, was conducted with a small sample size and short period of time. In [21], while the study included a large number of participants (925), the study itself was restricted to sending two text messages per week to encourage behaviour modification. It is unclear how the messages were designed and if they were tailored. The results are also silent on user acceptance. Donaldson [11] presented a mobile application through which users sent SMS bi-weekly and received feedback from practitioners. The study was limited in time and conducted with a very small sample size. The level of sophistication of the technology usage was also very low and limited to SMS exchange. Jen [22] presents results on the usability of a developed mobile weight management services in a virtual community. The application was tested and validated by about 400 college students. It will be interesting to follow the progress of this research, how it will be experimented with real participants and what impact it will have. In the following section, we describe the mobile application that was developed to address the discussed shortcomings in the literature. 3. Mobile Application Design To implement the three strategies described in the introduction, three interfaces were designed: A messaging system, a personal goal achievement system, and a group goal achievement system. The following describes a sample implementation walk-through for the three interfaces. Fig. 1 shows, from left to right, the home interface as a portal to subsequent interfaces (messaging, goals setting, group support and a progress bar). Clicking the “Read Messages” icon, leads to the “received messages” interface. The application checks the phone’s message database for new messages when loaded for the first time, or when a user clicks the home tab. The application searches for messages sent from a specific phone number, associated with the
841
842
Selma Limam Mansar et al. / Procedia Technology 5 (2012) 839 – 848
online SMS system and lists them in the ‘reading messages’ interface. The messages are sorted chronologically. A “write a message” button at the bottom of the ‘received messages’ interface directs to the ‘writing message’ interface. The latter uses the same top navigation tabs as the previous interfaces for consistency. The message writing box is invoked from the Android messaging API and the recipient’s number, which is the dietician or the SMS system’s number, is already linked with the current message. The user can now type the message and upon clicking send, the Android messaging API will send the message for our application to the number hard-coded into the application. If mobile data or Wi-Fi is enabled on the device, the message can also be sent to the research team’s database for storage and for the dietician to take action if the participant is seeking help. The user is given a success message once the message is sent and then returned to the first interface, where the application will again check for any new messages as described earlier. Fig. 2 describes the goals settings interfaces accessed through the “My Goals” icon. The left and center images reflect the ‘personal goal achievement setting’. Patients set up goals with a dietician during an initial health assessment. The goals are reflected in the mobile application. The share routine option is a way to manage up-dates towards personal goals. The third interface, on the right hand side, shows the participant’s sub-goals. The sub-goals, by default, are 0% completed. However, the application pulls data regarding the goals from the data-base every time the share routine interface is accessed. If there has been progress previously recorded by the user, the sliders are adjusted accordingly to reflect the progress. The sliders are constrained to five options; 0, 25, 50, 75 and 100% completion to maintain consistency. The user can use the slider to reflect the current progress of the goals. Since the application fetches progress data every time the interface is accessed, it allows the user to easily gauge their progress and update and to even make a correction in case a wrong value was entered by mistake. Clicking the save button at the bottom of the screen saves the progress values in the database as well as sends the same data to the online server to track progress. The progress bar on the home screen is simply a visual tool to communicate the participant’s current progress. The bar is also updated every time the home interface is accessed. After updating the goals on the share routine page, the user is directed to the home screen (see Fig. 1, far left image) and the application pulls the updated goal progress values to compute and draw the progress bar. The progress bar is computed simply by taking an average of all the individual values of the personal goal progress.
Fig. 1. A sample of the mobile application interface before usability testing: Home and Messaging interfaces.
From the home interface (Fig. 1, far left image), one can access the group progress, the third strategy described in the introduction, through a virtual social support network “Group Buzz”. The messaging and share
843
Selma Limam Mansar et al. / Procedia Technology 5 (2012) 839 – 848
routine icons serve the same purpose and behave the same way as described in the previous interfaces. Users here work towards achieving a collective group goal, fostering the power of group dynamics. The group progress is based on the aggregate score of all group members’ individual goals. Clicking the “group buzz” would display an interface similar to the read messages interface described above. The difference being that instead of listing direct messages sent by the user’s dietician, the interface lists all the messages posted by the members of the user’s group. These messages are online messages only and are not SMS. The messages are stored on the online database and are tagged with a group number. This is to ensure that only members of the group assigned the same tag numbers are able to post, share and read messages. When a user sends a message to the group, a participant’s assigned nickname, the group tag number and the timestamp is associated with that message. This makes it easy to filter messages according to groups and to display messages with the sender’s nickname and the date-time. Groups can send and receive messages only from their group members and cannot send or view another group member’s messages. However, the application is designed such that the research team or care givers can access all group messages so as to administer and moderate possibly inappropriate messages. 4. Mobile Application Usability Testing Experiment So far we have described the literature review as well as strategies to map behavior theories to a mobile application. We have described and implemented a mobile application that reflects our study. In this section, we describe how we tested the acceptability of such an application. The usability test was conducted with seven participants drawn from an accessible pool of young female adults.
Fig. 2. A sample of the mobile application interface before usability testing: Goals setting
We chose native speakers of Arabic and one gender to avoid biases. Before handing the application to participants, users were asked to test the smart phones pre-loaded with the Android application, by performing tasks relevant to each interface. For example, participants were asked to update a pre-defined sample goal regarding physical activity and food intake by navigating through the application home interface. Participants’ time on task for each task were recorded along with recording verbal protocols, Ericsson and Simon [23]. The times on task, as well as the verbal comments, were recorded for qualitative analysis. After completing the tasks, participants received a paper-based mobile usability questionnaire based on work by Ryu et al. [24] and Jumisko-Pyykkö et al. [25]. This usability study aimed at assessing immediate flaws with the three interface designs of the application. The table below summarizes the results gathered from the talk aloud protocols and
844
Selma Limam Mansar et al. / Procedia Technology 5 (2012) 839 – 848
the post-test survey. It indicates the issues identified by the users and their quantitative evaluation of usability (by allocating a score of 1, very low, to 5, high) for each tested feature. The table introduces the set of actions recommended as a result of the usability testing. Table 1. Usability Testing Results Tested features
Identified issues
Recommended actions
Mode
Users had difficulties identifying how to access the menu. Android phones include the menu button outside the apps.
Should include a send button
3
Slider
Users needed assistance when using the slider.
Should add a slider that shows percentage or numbers from one to four. The slider should also display a different colour when sliding.
3
Adding new goal
Goal was lost because user did not proceed to save.
There should be a dialogue box popping out to confirm/remind about saving.
3
Social networking Interface Group buzz
In addition to the built in menu button.
Goals Interface
User wasn’t sure about how to update the personal goals. Check personal progress
User could not understand how the progress was measured. Did a full bar indicate successful progress or no progress at all?
Include labels under the progress bar and have a different fill in color for the progress bar.
3
Update personal goal
User didn’t quite understand the purpose of the progress bar.
Add a motivational message under the progress bar.
2
User could not navigate back to the home screen
Change the text on the “save” button to “proceed to sub-goals”
Navigate and read a message
User couldn’t open read messages when pressing on the icon image.
Increase the icon canvas surface area to increase the touch interaction.
3
Reaching the homepage
The authors had to tell some users to keep using the back button to reach the home screen.
Users are not that familiar with the android OS. It is expected that a familiar user will know how to operate.
3
Navigation
Online Survey Meaningfulness of symbols, item labels and icons
Users didn’t find the icons culturally or contextually appropriate.
No action needed
4
Icons and symbols need to be redesigned.
3
Selma Limam Mansar et al. / Procedia Technology 5 (2012) 839 – 848
5. Mobile Application Localization Localization is defined by Shen et al. [26] as “the process of adapting a program for a specific international market”. A localized product allows users “to concentrate on exploiting the software in their own language and appropriate cultural context” [27]. After conducting the usability testing, several steps were taken to localize the mobile application. Localization means first and foremost adapting to the most local language. However, language alone is not enough. The literature emphasizes various dimensions: Shen et al. [26] emphasize Culturalisation, Metaphors, Semiotics (iconic design), Aesthetics (color, typography…) and Navigational flow (direction of visual scanning). Sandrini [28] emphasizes translation, Shared ‘Locale’: shared language, writing system, and cultural adaptation of translated words. We have redesigned the application to take into account the mentioned aspects of localization. The table be-low summarizes how localization was approached for the mobile application. While the results apply to the specific case of our mobile application, they nevertheless provide guidelines for developers of mobile applications in Arabic. It is worth noting that on the app store, and using key words relevant to weight loss, obesity and healthy lifestyle, only two mobile applications were found in Arabic for 400 applications scanned. Table 2. Actions for localization of the mobile application. Localization element
Definition
Actions taken
Iconization – Semiotics
Visual communication with icons.
Changed the icons to become more self-explanatory (although the text is still kept)
Culturalisation
Promoting a local
Characters were introduced dressed up in traditional clothing
Identity, highlighting culture and traditions. Metaphors
According to Shen [26], metaphors should be rich, suitable, fun and alluring, original, adaptable and transferrable.
“Wazni” the application’s name means “Weight” in English and is depicted inside a series of Arab style doors. The doors symbolize a hub or portal access to weight management. Doors also symbolize an entry to a private world, hence reassuring that the patient’s information and exchanges are private and the portal is personal.
Aesthetics/Color
Colors that are adapted to local culture
Screening local main hospitals and other health organizations websites, white and blue (or green) appeared to be recurring colors. A white and blue color palette was hence chosen.
Aesthetics/
Arranging type to make the language visible
We chose a Simplified Fixed Arabic Font type. The study usability conducted by Al-Wabil and George [29], recommended such a font over Traditional Arabic fonts. The former fonts were found to be faster and more accurate to read.
Navigating between interfaces and reading/visually perceiving information across one interface
Arabic reads from right to left. The menus and positioning of the icons was adjusted accordingly.
Typography
Navigation flow
Also, the goals slider was adjusted to slide from right (minimum value) to left (maximum value) According to George et al. [30], the navigation patterns in Arabic are similar to those in English, except for reading right to left and for banners blindness (minimal in Arabic).
845
846
Selma Limam Mansar et al. / Procedia Technology 5 (2012) 839 – 848
Translation
Displayed text and communicated in the most common language (Modern Standard Arabic and in the local dialect when appropriate)
Cultural adaptation of translated words
The parallel back translation method was used to ensure the translated words were faithful to the original meaning. A native Arabic speaker translated the text from English to Arabic and another Native speaker translated the Arabic text back to English. Where discrepancies in the original English text and the one obtained after the parallel back translation were found, parallel-back translation was reapplied.
As a result of the usability testing and the guidelines from the localisation study, the mobile application was improved as follows:
Fig. 3. A culturally appropriate mobile application for sustainable weight loss
6. Future Work The mobile application is now ready for testing with participants. There are two essential elements that need completion to insure a good usage of the applications: proper goals setting and appropriate messages to be exchanged between the dietician and the participants. Goals need to be positive, measurable and attainable in short timespans. The dietician will need to embrace this strategy. To prepare the database of messages, we
Selma Limam Mansar et al. / Procedia Technology 5 (2012) 839 – 848
plan to run a survey on the community eating and exercise habits before inviting participants to the study. The (anonymous) survey results will allow the researchers to better understand current behaviors and hence, design well targeted, relevant and meaningful messages. 12 participants will be invited to participate: 6 will be using the mobile application, and 6 will be invited to follow a traditional weight loss plan with targeted goals and act as a control group. Participants will engage in three weeks of experimentation, each week corresponding to one feature of the application. The analysis of the results will allow for an improvement of the application, but will not provide confidence on the health side. A second step would be to test it on a broader scale (with at least five weeks for each feature). 7. Conclusion We have presented a mobile application that aims at supporting sustainable weight loss by leveraging established behavior change theories. Three interfaces were designed and implemented: A messaging system, a personal goal achievement system and a group goal achievement system. The application was validated through a usability testing experiment. Seven young female adults, native Arabic speakers, tested the Android application in the smart phones and were asked to perform tasks relevant to each interface. Following the usability testing, some recommendations were made for improvement. We adapted and applied Shen at al. [26] and Sandrini [28] approaches to localization (designed initially for websites) to improve the mobile application for the Middle Eastern context. In future work, we will be conducting a five weeks experiment on volunteering participants to assess the effectiveness of the mobile application on behavior change. We will also explore research pertaining to the usage of Arabic language and customization of mobile applications. Another extension and a supplement for the application could be to connect it to a web portal, accessible by the caregivers as well as the users. The web portal can reflect the data collected through the mobile application (such as frequency of messaging, charts and statistics around goals attainment and caregiver’s follow up strategies with the participants).
Acknowledgements This publication was made possible by a grant from the Qatar National Research Fund under its Undergraduate Research Experience Program award number UREP11-144-1-025. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the Qatar National Research Fund.
References [1] Limam Mansar, S., Kekre S. A Founding Framework for addressing Obesity in Qatar using Mobile Technologies. Communications in Computer and Information Science 2000, 221, 3, pp. 402-412. [2] Stroebe, W. Dieting, overweight, and obesity: Self-regulation in food-rich environment. Washington, DC: American Psychological Association Press, 2008. [3] Stroebe, W., et al. Why dieters fail: Testing the foal conflict model of eating. Journal of Experimental Social Psychology 2008, 44, pp. 26-36. [4] Limam Mansar, S., Brunstein. A., Jariwala, S., Brunstein, J. Addressing Obesity Using A Mobile Application: An Experiment Design. e-Health 2012, Accepted for IADIS Multi Conference on Computer Science and Information Systems (MCCSIS 2012). [5] Huang, A. et al. The effects on saturated fat purchases of providing Internet shoppers with purchase-specific dietary advice: a randomized trial. PLoS Clinical Trials, 2006; 1:e2. [6] Peterson, S. et al. Positive changes in perception and selections of healthful foods by college students after a short-term point-ofselection intervention at a dining hall. Journal of American College Health 2010, 58, pp. 425-431.
847
848
Selma Limam Mansar et al. / Procedia Technology 5 (2012) 839 – 848
[7] Frenn, M., et al. Changing the tide: an Internet/video exercise and low-fat diet intervention with middle school students. Applied Nursing Research, 2005; 18:13–21. [8] Webb, T., et al. Using the Internet to promote health behavior change: A systematic review and meta-analysis of the impact of theoretical basis, use of behavior change techniques, and mode of delivery on efficacy. Journal of Medical Internet Research 2010, 12, e4. doi: 10.2196/jmir.1376. [9] Patrick, K. et al. A text message based intervention for weight loss: Randomized controlled trials. Journal of Medical Internet Research 2009, II, e l. [10] Gold, B. C. et al. Weight loss on the web: a pilot study comparing a structured behavioral intervention to a commercial program. Obesity 2007, 15, pp. 155–64. [11] Donaldson, E. L., and Fallows, S. A text message-based weight management intervention for overweight adults. Journal of Human Nutrition and Dietetics 2011, 24, 4, pp. 385–386. [12] Burnett, K., Taylor, C., & Agras, W. Ambulatory computer-assisted therapy for obesity: A new frontier for behavior therapy. Journal of Consulting and Clinical Psychology 1985, 53, 698-703. [13] Mattila, E. et al. Use of a mobile phone diary for observing weight management and related behaviours. Journal of Telemedicine and Telecare 2010. 16, pp. 260-264. [14] Goulis, D. G. et al. Effectiveness of home centered care through telemedicine applications for overweight and obese patients: a randomized controlled trial. International Journal of Obesity and Related Metabolic Disorders 2004, 28 (11), pp. 1391-1398. 31. [15] Block G, Block T, Wakimoto P, Block, C.H. Demonstration of an e-mailed worksite nutrition intervention program. Preventing Chronic Disease 2004; 1:A06. [16] Woolford. S. J., et al. Tailored mobile phone text messages as an adjunct to obesity treatment for adolescents. Journal of Telemedicne and Telecare 2010, 16:458-461. [17] Salvy, S. J. et al. Effects of social contexts on overweight and normal-weight children's food intake. Physiology & Behavior 2007, 92, pp. 840-846. [18] Salvy, S. J. et al. Peer Influence on Children's Physical Activity: An Experience Sampling Study. Journal of Pediatric Psychology 2008, 33 (1), pp. 39-49. [19] Lee, W. et al. Evaluation of a mobile-phone based diet game for weight control. Journal of Telemedicine and Telecare 2010, 16, pp. 270-275. [20] Kornman, K.P. et al. Electronic therapeutic contact for adolescent weight management: The Loozit Study. Telemedicine and ehealth, July/August 2010, pp. 678-685. [21] Joo, N-S. et al. Cost effectiveness of a community-based obesity control program. Journal of Telemedicine and Telecare 2010, Vol. 16, pp.63-67. [22] Jen, W-J. The adoption of mobile weight management services in a virtual community: the perspective of college students. Telemedicine and e-health, May 2010, pp. 490-497. [23] Ericsson, K.A., & Simon, H.A. Protocol analysis: Verbal reports as data. MIT Press, Cambridge, MA, 1993. [24] Ryu, Y. S. & Smith-Jackson, T. L. Reliability and Validity of the Mobile Phone Usability Questionnaire (MPUQ). Journal of Usability Studies 2006, Vol. 2, pp. 39-53. [25] Jumisko-Pyykkö, S. and Vainio, T. Framing the Context of Use for Mobile HCI. International Journal of Mobile-Human-ComputerInteraction (IJMHCI) 2010, 2, 1. [26] Shen, S-T., Woolley, M., Prior, S. Towards culture-centered design. Interacting with Computers 2006, 18, pp. 820-852. [27] Uren, E., Howard, R., Perinotti, T. Software Internationalization and Localization: An Introduction, vol. xx. Van Nostrand Reinhold, New York, p. 300, 1993. [28] Sandrini P. Website Localization and Translation. EU-High-Level Scientific Conference Series MuTra 2005 – Challenges of Multidimensional Translation: Conference Proceedings; 2005. Retrieved 24 June 2009 from www.euroconferences.info/proceedings/2005_Proceedings/2005_Sandrini_Peter.pdf [29] Al-Wabil, A, George, R.P. An eye tracking study of Arabic typography readability. Proceedings of the IADIS International Conference on Interfaces and Human Computer Interaction. Freiburg, Germany 26-30 July 2010. Edited by Katherine Blashki, 2010. ISBN (Book): 978-972-8939-18-2. [30] George, R.P., Anwar, R, Jeyasekhar, S. Visual Reading Patterns on Arabic Interfaces: Insights from Eye Tracking. Journal of Computing, Volume 3, Issue 11, November 2011, ISSN 2151-9617.
