therapy for naming deficits in aphasia

This article was downloaded by: [UQ Library] On: 05 November 2014, At: 23:27 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Aphasiology Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/paph20

Behavioural and neural changes after a “choice” therapy for naming deficits in aphasia: preliminary findings Carol Leonard cfg

abc

b

de

, Laura Laird , Hana Burianová , Simon eh

b

bci

Graham , Cheryl Grady , Tijana Simic & Elizabeth Rochon a

Audiology and Speech-Language Pathology Program, School of Rehabilitation Sciences, University of Ottawa, Ottawa, Canada b

Department of Speech-Language Pathology, University of Toronto, Toronto, Canada c

Heart and Stroke Foundation Canadian Partnership for Stroke Recovery, Ottawa, Canada d

Centre for Advanced Imaging, University of Queensland, Brisbane, Australia e

Rotman Research Institute, Baycrest, Toronto, Canada

f

Physical Sciences, Sunnybrook Research Institute, Toronto, Canada g

Department of Medical Biophysics, University of Toronto, Toronto, Canada h

Department of Psychiatry and Psychology, University of Toronto, Toronto, Canada i

Toronto Rehabilitation Institute, Toronto, Canada Published online: 04 Nov 2014.

To cite this article: Carol Leonard, Laura Laird, Hana Burianová, Simon Graham, Cheryl Grady, Tijana Simic & Elizabeth Rochon (2014): Behavioural and neural changes after a “choice” therapy for naming deficits in aphasia: preliminary findings, Aphasiology, DOI: 10.1080/02687038.2014.971099 To link to this article: http://dx.doi.org/10.1080/02687038.2014.971099

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to

the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.

Downloaded by [UQ Library] at 23:27 05 November 2014

This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/termsand-conditions

Aphasiology, 2014 http://dx.doi.org/10.1080/02687038.2014.971099

REPORT Behavioural and neural changes after a “choice” therapy for naming deficits in aphasia: preliminary findings


Carol Leonarda,b,c*, Laura Lairdb, Hana Burianovád,e, Simon Grahamc,f,g, Cheryl Gradye,h, Tijana Simicb and Elizabeth Rochonb,c,i a Audiology and Speech-Language Pathology Program, School of Rehabilitation Sciences, University of Ottawa, Ottawa, Canada; bDepartment of Speech-Language Pathology, University of Toronto, Toronto, Canada; cHeart and Stroke Foundation Canadian Partnership for Stroke Recovery, Ottawa, Canada; dCentre for Advanced Imaging, University of Queensland, Brisbane, Australia; eRotman Research Institute, Baycrest, Toronto, Canada; fPhysical Sciences, Sunnybrook Research Institute, Toronto, Canada; gDepartment of Medical Biophysics, University of Toronto, Toronto, Canada; hDepartment of Psychiatry and Psychology, University of Toronto, Toronto, Canada; iToronto Rehabilitation Institute, Toronto, Canada

(Received 29 May 2014; accepted 24 September 2014) Background: Anomia, difficulty producing words, is a pervasive symptom of many individuals with aphasia. We have developed a treatment for naming deficits—the Phonological Components Analysis (PCA) protocol—that has proven efficacious in improving word-finding abilities for individuals with post-stroke aphasia. Aims: The aim of this investigation is to present preliminary findings exploring the potential influence of choice—that is the active engagement of a participant in therapy —on our PCA treatment. Methods & Procedures: Five individuals with aphasia were treated in one of two conditions—Choice or No Choice. Potential changes in neural activation as a function of the treatment were also investigated. Two individuals (one from each condition) underwent functional MRI (fMRI) pre- and post-therapy. Outcomes & Results: All the individuals demonstrated a significant treatment effect immediately post-treatment and at a 4-week follow-up and four of the five participants at an 8-week follow-up. Three also demonstrated generalisation to untrained items. Unfortunately, no clear-cut patterns emerged to allow us to make claims about the influence of choice, per se, on the behavioural manifestations of improved naming. Interestingly, the participant from the Choice condition showed neural activation changes post-treatment in frontal and parietal regions that were not evident for the participant in the No Choice condition. Moreover, these changes were accompanied by a larger treatment effect for that individual and generalisation to a novel naming task. Conclusion: The efficacy of PCA treatment for naming deficits is further supported. In addition, the neuroimaging data suggest the possibility that active engagement of an individual in his/her therapy (in this case choosing phonological attributes of a target word) may exercise executive functions important for success in treating anomia. Also, continued exploration of task factors that may promote even better treatment effects using this protocol is warranted, as is continued investigation of the neural underpinnings associated with treatment effects. Keywords: stroke; aphasia; anomia; rehabilitation; neuroimaging; neuroplasticity

This report presents preliminary data exploring the potential influence of a particular task factor, the element of choice, on our phonologically based treatment for anomia—the *Corresponding author. Email: [email protected] © 2014 Taylor & Francis


2

C. Leonard et al.

Phonological Components Analysis (PCA) protocol (Leonard, Rochon, & Laird, 2008). The element of choice (or active engagement of a participant in therapy) has been raised as a potential factor that leads to better treatment effects. Hickin, Best, Herbert, Howard, and Osborne (2002) suggest that deeper processing during therapy may be promoted by an individual’s engagement in the task through active choice and that treatment effects may, in turn, prove to be longer lasting. In addition, data addressing the neural changes that occur as a function of the therapy are presented, following upon our previous work addressing this topic (Rochon et al., 2010). PCA therapy is a modified version of Coelho, McHugh, and Boyle’s (2000) protocol for semantic features therapy. Participants are asked to name a picture (e.g., bed) and identify five phonological components associated with that target word (e.g., rhyming word → shed, first sound → /b/, last sound → /d/). In Leonard et al. (2008), we demonstrated the efficacy of our PCA treatment for anomia in individuals with aphasia. Using a single-subject multiple baseline across behaviours design, we found that the naming performance of seven of the 10 participants improved, with very good maintenance of gains at 4-week follow-up, and evidence of generalisation to untrained stimuli on a standardised naming test. While these results were encouraging, there were three individuals in our study for whom this treatment was not effective. It is still unclear why treatment is successful for some individuals and not for others. We suggested that our results could be accounted for in part by the notion, advanced by Hickin et al. (2002), that active engagement on the part of the participant in his/her therapy is necessary to produce longer lasting effects. Phonological components (i.e., prompts) in our treatment were not simply supplied to the participants, but rather participants were provided with the opportunity to generate and/or choose their own components for a given target word. Interestingly, five of the seven participants for whom treatment was successful were able to generate the components independently at least 50% of the time (averaged across the three lists), as compared to 40% or less in the group for whom treatment was not successful. Many others have emphasised the importance of active participation to treatment success for word-finding deficits in aphasia (e.g., Francis, Clark, & Humphreys, 2002; McNeil, Small, Masterson, & Fossett, 1995; Spencer et al., 2000; Wiegel-Crump & Koenigsknecht, 1973). In McNeil et al. (1995), for example, treatment focused on having the participant provide synonyms and antonyms for a word to be named and Spencer et al. (2000) had their patient respond to questions about semantic and phonological attributes of the target words. Both found improved naming post-treatment. Francis et al. (2002) developed a treatment termed “circumlocution-induced naming” in which the patient was encouraged to talk around the target to be named until she was able to produce it. They found improved naming on treated as well as untreated words and suggested that success may be attributed to the fact that this type of therapy encourages the patient to be actively involved in “exercising” lexical connections. Indeed, the relationship between successful anomia therapy and cognitive abilities/ executive function (such as attention and problem-solving) has also been established (e.g., Fillingham, Sage, & Lambon Ralph, 2005a, 2005b, 2006; Lambon Ralph, Snell, Fillingham, Conroy, & Sage, 2010; Yeung, Law, & Yau, 2009). In this vein, one might argue that active engagement (through the element of choice) of a patient in his/her therapy also “exercises” and strengthens the cognitive abilities important for treatment success. As noted earlier, in addition to tracking changes in naming performance as a function of PCA treatment, we also investigated the neural processing characteristics associated with improved naming (Rochon et al., 2010). Using functional MRI (fMRI), we compared


Aphasiology

3

neural changes related to completing a phonological task and a semantic task before and after treatment. Two of the treated patients in Leonard et al. (2008) and two untreated patients were scanned. In both treated patients whose naming had improved post-treatment, we found changes associated with semantic processing in the left inferior frontal gyrus (LIFG) and left middle temporal gyrus, areas typically associated with semantic processing in healthy participants (e.g., Perani et al., 2003). In addition, consistent with some previous studies (e.g., Cornelissen et al., 2003; Meinzer et al., 2008), we found increased left hemisphere (LH) processing compared to right hemisphere (RH) processing post-treatment. This finding supports recent work by Fridriksson, Richardson, Baker, and Rorden (2011) who found, using transcranial direct current stimulation of the LH in addition to behavioural anomia therapy, evidence to suggest that LH activation is extremely important to the recovery of naming abilities. We also found that activation posttreatment during the semantic task involved areas more typically associated with phonological processing; specifically the left supra-marginal gyrus and inferior parietal regions (e.g., Lurito, Kareken, Lowe, Chen, & Mathews, 2000). This finding led to the speculation that our phonologically based treatment strengthened the connections between the lexical and phonological processing levels involved in naming. In addition, we suggested that perhaps, while our treatment focused on the phonological aspects of target words, semantic processing was also activated since pictures were used as stimuli and correct production necessarily involved the activation of both semantic and phonological elements. The present investigation represents a preliminary inquiry into the factor of active engagement or choice associated with the success of PCA treatment on naming performance in individuals with aphasia and associated neural underpinnings. In particular, we compared PCA treatment in two conditions: Choice and No Choice. The conditions differed as to whether the participant was allowed to choose the phonological components of the target word or whether he/she was simply provided with the response. We hypothesised that while the two conditions may result in similar immediate treatment effects, the Choice condition would lead to better effects of maintenance of targeted items and generalisation to untrained lexical items. With respect to the neuroimaging results, we hypothesised that post-therapy, for individuals in both groups (Choice/No Choice) when performing both the phonological and semantic tasks, there would be a shift from greater RH processing to greater LH processing associated with improved performance in naming and more LH peri-lesional activation. Moreover, because the therapy specifically targets phonological processing, greater activation in the left supramarginal gyrus post- versus pre-therapy was expected. Associated increased activation by treated patients in the anterior LIFG and middle temporal areas during the semantic judgement task (and accompanying RH to LH shift) would provide evidence of the influence of a phonological feature-based therapy on semantic processing. In addition, based on the literature (e.g., Turner & Levine, 2004) emphasising the importance of frontal regions to executive processing and active engagement, increased neural activation in frontal regions post-treatment was expected for individuals in the Choice condition.

Method for treatment phase Participants Five individuals with aphasia participated in this study. Participants were randomly assigned to one of two treatment conditions: Choice—comprised of two men and one


4

C. Leonard et al.

woman (mean age: 81.3 years; mean level of education: 15 years); or No Choice condition —comprised of two men (mean age: 71.5 years; mean level of education: 9.5 years). All participants were right-handed individuals who had experienced a single left-hemisphere stroke and were at least 1-year post-onset at the time of enrolment. Four were native English speakers and one (P1) was educated in English at university (undergraduate and Master’s levels) and worked in English as an engineer for many years. Classification of aphasia was based on the results of the Boston Diagnostic Aphasia Exam (Goodglass, Kaplan, & Barresi, 2001) and revealed four individuals with moderate-to-severe Broca’s aphasia (three in the Choice condition) and one with mild-to-moderate Wernicke’s aphasia. Inclusionary criteria included a naming impairment defined by less than 75% correct (with a minimum of 10% accuracy) on the Boston Naming Test (Goodglass et al., 2001) and/or preservation of some oral reading skills as measured by at least 20% accuracy on the Oral Reading Subtest (31) of the Psycholinguistic Assessments of Language Processing in Aphasia (PALPA) (Kay, Lesser, & Coltheart, 1992). In addition, all the participants had (corrected) normal vision and visual perceptual abilities within normal limits as defined by performance on the Birmingham Object Recognition Battery (Riddoch & Humphreys, 1993). Hearing was screened and found to be within normal limits for all. As well, no participant presented with a significant apraxia of speech, and none was receiving formal speech–language therapy at the time of testing. Other exclusion criteria included a known history of drug or alcohol abuse, a history of major psychiatric illness and/or other neurological illness. See Table 1 for a summary of participant information. Treatment The treatment stimuli were the same as those used in Leonard et al. (2008) and consisted of 105 coloured photographs of nouns. Baseline testing Three baseline sessions were conducted. At each session, participants were shown all of the pictures to name. Words that were in error on at least two of the three sessions served as the potential pool of words to be targeted in therapy. Participants were asked to choose from this pool the items that they would like to have treated. Of those, 30 words1 were chosen and divided into two lists that were equated as much as possible by category, word frequency, and number of syllables. Responses provided by the participants during the baseline sessions were also rated on a 4-point scale (0 = incorrect; 1 = semantic or phonological paraphasia; 2 = semantic or phonological paraphasia with immediate self-correction of target word; 3 = correct) (Huber, Poek, Weniger, & Willmes, 1983). Also, as in Leonard et al. (2008), to ensure some success during treatment sessions, up to 10 additional words that the participant could name were included in the lists, but not considered part of the baseline. Treatment As noted, our PCA therapy is a modified version of Coelho et al.’s (2000) protocol for semantic features therapy. A picture of the target word was presented in the centre of a chart and participants were asked to name it. Next, irrespective of the patient’s ability to name the target, he/she was required to identify five phonological components of the

77.4

Choice Choice Choice No Choice No Choice

Mean

1 2 3 4 5

12.8

16 14 15 12 7

Education (years)

b

1.65

3 1 1 1.75 1.5

Post-stroke (years) M M F M M

Sex L L L L L

MCA MCA MCA CVA temporal & parietal infarct

Etiology

a

Broca’s Broca’s Broca’s Broca’s Wernicke’s

Aphasia typeb 2 2 1 2 3

Severity ratingb

35 17 0 13 20

BNTc (% correct)

56 46 46 36 43

PALPA 31d (% correct)

Note: According to CT scan and/or neurological reports; Based on Boston Diagnostic Aphasia Examination (BDAE), Goodglass et al. (2001); cBoston Naming Test (BNT), Goodglass et al. (2001); dOral Reading Subtest 31—Psycholinguistic Assessments of Language Processing in Aphasia (PALPA), Kay et al. (1992).

a

74 81 89 64 79

Participant Condition

Participant characteristics.

Age (years)

Table 1.


Aphasiology 5

6

C. Leonard et al.

target word (rhyming words: “What does this rhyme with?”; first sound: “What sound does it start with?”; first sound associates: “What other word starts with the same sound?”; final sound: “What sound does it end with?”; number of syllables: “How many beats does the word have?”), guided by the use of a chart. Identification of the components proceeded differently, depending on the treatment condition.


Choice condition. For each of the five components to be identified, the participant was asked to provide a response. If the participant could not spontaneously provide a response, then he/she was asked to choose one from a list of up to four correct responses. The choices were visually presented on a card and read aloud by the examiner. No choice condition. For each of the five components identified, the participant was provided with the response.2 For each phonological component, the examiner wrote on the chart the component that the participant chose (Choice condition) or was provided (No Choice condition). Note that the target word was never written down. Once all the components were reviewed, the examiner asked the patient to name the target again. If the target was named correctly, the examiner provided appropriate positive feedback, “Yes, that’s right. It’s a ____.” When the response was incorrect, the examiner said the correct response and asked the participant to repeat it. The next step in the treatment protocol involved having the examiner review all of the phonological components with the participant and then asking him/her to name the target again. Feedback at this stage for correct and incorrect responses was as described above. This procedure allowed the examiner to provide the name of the target word twice, irrespective of whether the participant was correct or incorrect in his/her original response. For an example of the PCA chart and a flowchart depicting the treatment protocol, see Leonard et al. (2008). Treatment schedule Treatment sessions (approximately 1 hour in duration) occurred three times a week for approximately 10 weeks (30 sessions in total). During each session, words in one of the two treatment lists were targeted. Target words were randomly presented at each session. Immediately following treatment and again at 4- and 8-week follow-up periods, participants were asked to name all treated items, and responses were rated on the 4-point scale. A rater blind to the purpose of the study scored these responses. Reliability Twenty per cent of the treatment sessions were videotaped, and all the sessions were audiotaped. Point-to-point agreement was used as a measure of reliability for the administration of treatment and the scoring of responses and found to be high (administration of therapy 100%; scoring as correct/incorrect 93%; scoring on 4-point scale 81% and 88% pre- and post-therapy, respectively (see Leonard et al. (2008) for a detailed description of the procedures used). Generalisation A comparison of performance on the Philadelphia Naming Test (PNT) (Roach, Schwartz, Martin, Grewal, & Brcher, 1996) pre- and post-therapy provided a measure of the generalisation of treatment effects to untreated items. Point-to-point agreement on scoring as correct/incorrect was 95% pre-therapy and 97% post-therapy.

Aphasiology

7

For each participant, the mean score (based on the 4-point scale; 0 = incorrect; 1 = semantic or phonological paraphasia; 2 = semantic or phonological paraphasia with immediate selfcorrection of target word; 3 = correct) for naming treated items at baseline was compared to naming performance immediately post-treatment and at 4- and 8-week follow-up periods3 using the Wilcoxon signed-rank test. Performance was significantly better (one-tailed) immediately post-treatment for P1 (Z = −2.95, p = .002), P2 (Z = −3.94, p < .001), P3 (Z = −3.42, p = .001), P4 (Z = −2.50, p = .006), and P5 (Z = −4.60, p < .001). Performance was also significantly better (one-tailed) at the 4-week follow-up for all the participants (P1: Z = −3.81, p < .001; P2: Z = −3.78, p < .001; P3: Z = −3.21, p = .001; P4: Z = −3.78, p < .001; P5: Z = −4.36, p < .001). At the 8-week follow-up period performance was significantly better (one-tailed) for P1 (Z = −3.61, p < .001), P2 (Z = −3.71, p < .001), P4 (Z = −3.97, p < .001), and P5 (Z = −4.29, p < .001), but not P3 (Z = −1.49, p = .068) (see Figure 1). For each participant, treatment effect sizes were calculated. Table 2 shows the mean effect size for each participant. Robey and Beeson (2005; in Beeson & Robey, 2006) suggest that for treatment targeting lexical retrieval an effect size of 10.1 is large, 7.0 is medium, and 4.0 is small. According to these criteria, two patients (P2 and P5) approached a small effect size.

3.00 2.50 Score on 4-point scale


Results of treatment phase

2.00

Baseline

1.50

Post-treatment 4-week follow-up

1.00

8-week follow-up

0.50 0.00

P1

P2

P3 Participant

P4

P5

Figure 1. Mean rating on a 4-point scale naming treated words per participant at four time points: Baseline, post-treatment, 4-week follow-up and 8-week follow-up.

Table 2. Participant P1 P2 P3 P4 P5

Mean effect size across treatment lists. Mean effect size 1.70 3.94a 1.85 1.53 3.28a

Note: aApproaches a small effect size according to Robey and Beeson (2005).

8

C. Leonard et al. Table 3.

Per cent correct on the PNT pre- and post-therapy.

Participant

Pre

Post

P1 P2 P3 P4 P5

67 63 7 33 45

63 75a 26a 38 72a


Notes: aDifference significant (p < .01, one-tailed), McNemar test. PNT— Philadelphia Naming Test (Roach et al., 1996).

Per cent correct on the PNT (Roach et al., 1996) was compared pre- and post-therapy using the McNemar test. A significant difference (p < .01, one-tailed) was found for P2, P3, and P5, with increased accuracy for all three patients post-treatment (see Table 3). Method for imaging phase Participants Two of the five participants in the treatment phase underwent neurological imaging. One participant (P2) was in the Choice condition and the other one (P4) was in the No Choice condition. fMRI activation tasks The two experimental tasks (semantic and phonological) and their corresponding control tasks were the same as those used in Rochon et al. (2010). In brief, a semantic judgement task, using the pictures from the Pyramids and Palm Trees Test (Howard & Patterson, 1992) was used as the experimental semantic task. The computer presented three pictures (one on top and two on the bottom), and participants were required to indicate, with a key press, which bottom picture was related in meaning to the one on top. The experimental phonological task was a rhyme judgement task using picture stimuli from a subtest of PALPA (Kay et al., 1992). The computer presented two pictures side by side, and the participant was required to indicate, with a key press, whether the words are rhymed or not. The computer automatically recorded reaction time (RT) and accuracy (see Rochon et al., 2010, for further details regarding the control tasks and baseline task). fMRI protocol For P2, scanning occurred 1 week prior to treatment and for P4 5 weeks. The second scans occurred 1 week post-treatment for P2 and 2 days post-treatment for P4. Scans were obtained using a 3.0 Tesla system (Signa Eclipse, GE Medical Systems, Milwaukee, WI). A T1-weighted volumetric anatomical MRI was obtained for each participant (124 axial slices, 1.4 mm, FOV = 22 cm). In addition, for each participant, T2*-weighted functional images were acquired using a spiral in/out pulse sequence (26 axial slices, 5 mm, TR = 2000 ms, TE = 30 ms, FOV = 20, 64 × 64 matrix). The blood oxygenation level-dependent effect was used to assess brain activation. A block design was used for the acquisition of the fMRI data. For both the semantic and the phonological tasks, six runs were used. Runs were divided into three blocks

Aphasiology

9

(experimental, baseline, and control), with each block containing four trials. The stimulus presentation rate was one per 8 s; so each block was 32 s and each run 96 s. The total duration for the presentation of the six runs was 576 s. The six runs used for the semantic task were presented together as were those comprising the phonological task (see Rochon et al., 2010, for example trials).


fMRI data analysis As in Rochon et al. (2010), fMRI images were preprocessed using the Analysis of Functional Neuroimages software (Cox, 1996). The images were motion-corrected, spatially coregistered, and smoothed. To further improve the signal-to-noise ratio in the patients’ images we used Independent Components Analysis (ICA) (Kochiyama et al., 2005). Each participant’s data set was analysed individually using Partial Least Squares (PLS) analysis (McIntosh & Lobaugh, 2004). PLS is a multivariate technique, which examines the covariance between activity in all brain voxels and experimental conditions, providing sets of mutually independent spatial patterns depicting brain regions that show the strongest relation to the contrasts across tasks. Using PLS, latent variables (LV), defined as cohesive patterns of neural activity associated with a task, were identified (the LV accounting for the most covariance is extracted first) across conditions and Scans 1 and 2. A permutation test (McIntosh, Bookstein, Haxby, & Grady, 1996) determined the significance of each LV and a bootstrap estimation of the standard errors determined the reliability of each LV (Efron & Tibshirani, 1986). Note that this analysis reveals the contrasts that account for the most covariance between the experimental tasks and brain activity: contrasts are not specified in advance.

Results of imaging phase Behavioural performance in the scanner Table 4 shows the accuracy (per cent correct) and RT data for P2 and P4 on the phonological experimental and baseline tasks. Table 5 shows the accuracy and RT data for P2 and P4 on the semantic experimental and baseline tasks. The McNemar change test was used to assess the difference in accuracy between Scans 1 and 2 on both the phonological and semantic tasks for each patient. Only one comparison was significant (p < .01; one-tailed), but it is not in the expected direction; at Scan 2 P4 was less accurate on the semantic task than at Scan 1. The Wilcoxon signed-rank test was used to examine the differences in RTs between Scans 1 and 2 on both the phonological and semantic tasks for each patient for correct responses only. Two comparisons were significant, both for P4. For both the phonological and the semantic tasks, P4 was significantly faster (Z = −1.78, p = .037; Z = −1.78, p = .038, respectively; one-tailed) at Scan 2 than at Scan 1. fMRI results PLS analyses were carried out comparing each participant’s activity during the phonological and semantic tasks across both the scans. For P2 comparison of the phonological to the baseline task yielded two significant LVs. LV1, accounting for 54% of the variance (p = .014), is largely accounted for by patterns of activation at Scan 1, before treatment. LV2, accounting for 37% of the variance (p = .046), represents changes in activation on this task at Scan 2 in left and right prefrontal

P2 P4

Patients

Table 4.

Mean (SD) Mean (SD)

3908 (949) 3339 (906)

RT

Scan 1

56 42

Accuracy 3970 (807) 2601 (613)

RT

Scan 2

Phonological experimental task

54 54

Accuracy

745 (343) 1047 (446)

RT

Scan 1

100 100

Accuracy

100 100

Accuracy

Scan 2

629 (267) 505 (211)

RT

Phonological baseline

Accuracy (per cent correct) and reaction time (RT, ms) data for P2 and P4 on the phonological and baseline tasks at Scans 1 and 2.


10 C. Leonard et al.

P2 P4

Patients

Table 5.

Mean (SD) Mean (SD)

4336 (1099) 3781 (1294)

RT

Scan 1

50 88

Accuracy 4060 (1377) 2775 (1133)

RT

Scan 2

Semantic experimental task

45 46

Accuracy 772 (418) 817 (408)

RT

Scan 1

83 83

Accuracy

83 83

Accuracy

Scan 2

764 (436) 755 (583)

RT

Semantic baseline

Accuracy (per cent correct) and reaction time (RT, ms) data for P2 and P4 on the semantic and baseline tasks at Scans 1 and 2.


Aphasiology 11


12

C. Leonard et al.

areas, thalamus, right middle occipital gyrus, and precuneus. Comparison of the semantic to the baseline task yielded a significant LV, which accounted for 80% of the variance (p < .001). P2 showed changes in activation on the semantic task after treatment (i.e., Scan 2) in left and right prefrontal regions, left cingulate gyrus, left middle temporal gyrus, and right insula. Figure 2 shows the design scores for both of these analyses with representative cortical activity maps and Tables 6 and 7 show significant activations on the phonological and semantic tasks, respectively. For P4, comparison of the phonological to the baseline task yielded one significant LV accounting for 68% of the variance (p = .002). As for P2, this effect appears to be due to patterns of activation at Scan 1, before treatment. Comparison of the semantic to the baseline task yielded a significant LV, which accounted for 79% of the variance (p < .001). In contrast to P2 where the effect occurred after treatment, this effect in P4 was evident both before and after treatment (i.e., no significant changes in activation seen after treatment). P4’s imaging data can be seen in Figure 3 and Tables 6 and 7. Discussion The purpose of the present investigation was to explore the potential influence of choice on the efficacy of our phonologically based treatment for anomia—PCA—and to identify associated neural underpinnings. Results indicated that all five participants showed improved naming performance post-treatment, irrespective of the treatment condition. Moreover, the treatment effect was maintained at 4- and 8-week follow-up periods, and generalisation of naming to untrained items was noted for three participants, two of whom were in the Choice condition and one in the No Choice condition. With this small number of participants, however, no clear-cut patterns emerged to suggest that the element of choice, per se, resulted in improved naming performance, nor that it particularly influenced the maintenance of treatment effects and generalisability to untrained items. With respect to the imaging data, while acknowledging the limitation inherent in the small sample size, the results, nonetheless, are somewhat more intriguing. Despite the fact that both participants showed improved naming after treatment, P2 (from the Choice condition) showed neural activation changes post-treatment in a number of areas that were not evident for P4 (No Choice condition). These changes were most notable in the left and right frontal regions. Interestingly, these changes in neural activation were accompanied by a larger treatment effect size for P2 compared to P4 and generalisation to a novel naming task (once again, not evidenced by P4). These findings speak to the issue of neuroplasticity associated with our treatment protocol and seem consistent with the results of Fillingham et al. (2005b) and others (e.g., Leśniak, Bak, Czepiel, Seniów, & Członkowska, 2008; Robertson & Murre, 1999; Turner & Levine, 2004) who have argued that frontal executive systems in particular are crucial for optimal gains in rehabilitation. With respect to our PCA treatment and the element of choice, we suggest that the active engagement of the participant in his/her therapy exercises the executive functions proposed by Sohlberg and Mateer (2001) of initiation, generative thinking and self-awareness. As with a phonemic fluency task (used to measure executive function) whereby a participant is required to generate as many words as he/she can think of that start with the same initial letter, our PCA therapy requires the participant to engage in similar processes (e.g., generate rhyming words and words that start with the same first sound of the target stimulus). Arguably, the engagement of these aspects of executive control during therapy may improve treatment outcomes.

Aphasiology

13

(a)

P2

Brain scores

150 100 50 0

Phonological

–50

Baseline

–100 –150

Scan 1

Scan 2

P2

Brain scores

150 100 50 0

Phonological

–50

Baseline

–100 –150 Scan 1

Scan 2

(c)

P2 250 Brain scores


(b)

150 50 Semantic

–50

Baseline

–150 –250

Scan 1

Scan 2

Figure 2. Results of the analysis comparing activation in the phonological and baseline tasks at Scans 1 and 2 for P2, as shown by the first latent variable (LV1) in Figure 2a and the second latent variable (LV2) in Figure 2b. Activations denoted in red represent brain regions that positively correlate with the phonological task and negatively correlate with the baseline task. Activations denoted in blue represent the brain regions that positively correlate with the baseline task and negatively with the phonological task. LV1 distinguishes the pattern of activation between these two tasks at Scan 1. LV2 distinguishes the pattern of activation between these two tasks at Scan 2. Figure 2c shows the results of the analysis comparing activation in the semantic and baseline tasks at Scans 1 and 2 for P2. This is shown by the first latent variable (LV1), which distinguishes the pattern of activation between these two tasks at Scan 2. Error bars denote 95% confidence intervals for the correlations calculated from the bootstrap procedure. [To view this figure in colour, please see the online version of this Journal.]

14 Table 6.

C. Leonard et al. Activations for P2 and P4 on the phonological task. MNI coordinates

Participant P2 (LV1) (LH) (RH)


P2 (LV2) (LH) (RH)

P4 (LV1) (LH)

Region

BA

X (mm)

Y (mm)

Z (mm)

Ratio

Inferior frontal gyrus Superior temporal gyrus Cingulate gyrus Middle occipital gyrus Orbitofrontal gyrus Inferior frontal gyrus Superior temporal gyrus

45 38 31 18 11 44 38

−60 −36 −20 28 20 52 32

36 4 −48 −80 44 12 4

0 −40 24 −4 −32 20 −36

7.6018 7.2574 7.0178 11.1553 10.518 8.9544 8.2954

Inferior frontal gyrus Middle and inferior frontal gyrus Medial frontal gyrus Thalamus Superior frontal gyrus Middle occipital gyrus Precuneus

47 10 10 8 19 7

−55 48 16 24 4 32 12

36 56 52 −24 56 −60 −40

−16 −4 −8 16 44 16 48

5.5315 19.5902 5.3883 4.2289 4.0466 3.7936 3.6545

Cuneus Insula Inferior frontal gyrus

17 13 44

−16 −36 −48

−76 4 4

8 0 16

14.9999 7.877 6.5783

Inferior frontal gyrus Superior frontal gyrus

45 10

64 12

24 68

16 −4

10.919 9.7921

(RH)

Notes: BA = Brodmann area; MNI = Montreal Neurological Institute; LH = left hemisphere; RH = right hemisphere. (P2) LV1: p = .014; 54% covariance; LV2: p = .046; 37% covariance; (P4) LV1: p = .002; 68% covariance; Positive ratios correspond to regions with positive salience on the LV. Negative ratios correspond to regions with negative salience on the LVs. X (right/left): negative values are in the LH; Y (anterior/posterior): negative values are posterior to the zero point (located at the anterior commissure); Z (superior/inferior): negative values are inferior to the plane defined by the anterior and posterior commissures. Ratio, salience/SE ratio from the bootstrap analysis, which is a measure of each voxel’s reliability.

The results of the present study are mostly similar to those of other studies and to our previous study (Rochon et al., 2010) regarding the areas of activation that were identified after successful treatment for anomia. However, the current results also differ somewhat from our previous study (Rochon et al., 2010) even though we used the PCA treatment and the same fMRI tasks. In our first study, we found changes in activation after treatment only on the semantic task and predominately in the LH; in the present investigation changes in activation were found on both the phonological and the semantic tasks and, for P2, changes on the phonological task were predominantly in the RH. Based on these findings one might speculate that the element of ‘choice’ that was manipulated in this study led to post-treatment changes in activation in the phonological task in addition to changes on the semantic task. It is also of interest to note that it has been suggested that the right inferior frontal gyrus may play a more important role in language processing with larger lesions and where there is less pronounced ipsilesional activity (e.g., Cappa, 2011; Meinzer, Harnish, Conway, & Crosson, 2011; Sebastian & Kiran, 2011). P2’s relatively large lesion and the fact that the phonological task recruited a relatively large amount of RH activity even before treatment (i.e., see the first significant

Aphasiology Table 7.

15

Activations for P2 and P4 on the semantic task. MNI coordinates

Participant P2 (LV2) (LH)


(RH) P4 (LV1) (LH) (RH)

Region

BA

X (mm)

Y (mm)

Z (mm)

Ratio

Parahippocampal gyrus Thalamus Cingulate gyrus Lingual gyrus Superior frontal gyrus Middle temporal gyrus Middle frontal gyrus Inferior frontal gyrus Insula Superior frontal gyrus Superior frontal gyrus Cingulate gyrus Inferior frontal gyrus Middle frontal gyrus Lingual gyrus Middle frontal gyrus

35

−24 −20 −12 −20 −8 −64 −32 60 36 16 −16 −20 −56 −28 8 20

−12 −16 −4 −68 −12 8 12 16 −20 8 44 8 12 28 −76 0

−24 16 36 4 64 −20 36 24 12 60 44 44 28 40 0 56

11.3868 7.7749 6.0772 5.7634 5.6704 5.2785 4.9332 12.542 5.5609 4.6445 7.0765 5.2219 5.1105 4.8539 13.8308 7.032

24 19 6 21 8 9 13 6 8 32 9 8 18 6

Notes: BA = Brodmann area; MNI = Montreal Neurological Institute; LH = left hemisphere; RH = right hemisphere. (P2) LV2: p < .001; 80% covariance; (P4) LV1: p < .001; 79% covariance: Positive ratios correspond to regions with positive salience on the LV. Negative ratios correspond to regions with negative salience on the LVs. X (right/left): negative values are in the LH; Y (anterior/posterior): negative values are posterior to the zero point (located at the anterior commissure); Z (superior/inferior): negative values are inferior to the plane defined by the anterior and posterior commissures. Ratio, salience/SE ratio from the bootstrap analysis, which is a measure of each voxel’s reliability.

LV for P2) are in keeping with these suggestions. In addition, rhyming has been shown to activate areas in the right inferior frontal gyrus (Calvert et al., 2000). This points to the importance of considering the influence of the task (see Cappa, 2011; Meinzer et al., 2011) on our findings as the task used here was not one of overt naming but of rhyme judgement. Lastly, the preponderance of changes in brain activation after therapy in the LH on the semantic task is in line with the importance that has been ascribed to increased LH activation for recovery in anomia, which has been well documented (e.g., Fridriksson, 2010; see Meinzer et al., 2008; Rochon et al., 2010). We can also speculate that improvement on the semantic task after treatment is compatible with the notion that PCA treatment, consistent with interactive activation models of lexical access (e.g., Foygel, 2000), exerts its effect by spreading activation from the phonological level to the semantic level, thereby boosting access to the semantic information necessary for word production (Leonard et al., 2008; Rochon et al., 2010; Van Hees, Angwin, McMahon, & Copland, 2013). There is yet another intriguing difference between the present results and our previous ones. In Rochon et al. (2010), both the treated patients who improved after PCA treatment showed changes in activation patterns after treatment. In the present study, while both the patients improved with PCA treatment, only one (P2) showed changes in activation after treatment. Although the reasons for this remain to be further explored, one possibility has to do with the magnitude of the treatment effects observed. The treatment effect size for P2 was 3.9, whereas for P4 it was 1.5. In Rochon et al. (2010), the treatment effect sizes

16

C. Leonard et al. (a)

P4

Brain scores

150 100 50 0

Phonological

–50

Baseline

–100 –150 Scan 1

Scan 2

P4 150 Brain scores


(b)

100 50 0

Semantic

–50

Baseline

–100 –150 Scan 1

Scan 2

Figure 3. Results of the analysis comparing activation in the phonological and baseline tasks at Scans 1 and 2 for P4, as shown by the first latent variable (LV1) in Figure 3a. Activations denoted in red represent brain regions that positively correlate with the phonological task and negatively correlate with the baseline task. Activations denoted in blue represent the brain regions that positively correlate with the baseline task and negatively with the phonological task. LV1 distinguishes the pattern of activation between these two tasks at Scan 1. Figure 3b shows results of the analysis comparing activation in the semantic and baseline tasks at Scans 1 and 2 for P4. This is shown by the first latent variable (LV1), which shows that the pattern of differences in activation between these two tasks is the same at Scans 1 and 2. Error bars denote 95% confidence intervals for the correlations calculated from the bootstrap procedure. [To view this figure in colour, please see the online version of this Journal.]

for the two patients who improved were comparable to P2’s (3.0 and 3.47), leading us to speculate that there may be a certain threshold of behavioural improvement that has to be exceeded in order for concomitant changes in neural activation patterns to be detected. In terms of the behavioural performance of the participants in the scanner, the results are not very revealing. For P2, no changes were found between Scans 1 and 2 for either the phonological or semantic task in terms of both accuracy and RT. For P4, accuracy on the semantic task was poorer at Scan 2 than Scan 1. However, this may be attributable to a speed/accuracy trade-off since RTs at Scan 2 were also significantly faster than at Scan 1. Some limitations of this study must be mentioned. First, the small sample size precludes clear conclusions regarding the task factor that was manipulated in this study,


Aphasiology

17

the element of “choice”. This and other task factors that may influence success in treatment for anomia remain to be further studied and elucidated. Second, the specific tasks used in the fMRI protocol no doubt influenced the findings. While the phonological and semantic judgement tasks employed here presumably reflected some of the underlying cognitive and linguistic processes involved in naming (Van Hees et al., 2013; see DeLeon et al., 2007), our results do not bear directly upon the task of overt naming per se. Although our use of a treatment approach targeted to a specific impairment allowed us to attempt to isolate the underlying cognitive operations involved in the treatment task (not often seen in larger, group studies), future work will include larger numbers of patients and an overt naming task. Many other aspects related to brain plasticity associated with our aphasia treatment remain to be further examined, such as the areas of activation associated with correct versus incorrect responses (e.g., Fridriksson, Bonilha, Baker, Moser, & Rorden, 2010; Postman-Caucheteux et al., 2010), treatment maintenance (Menke et al., 2009; Vitali et al., 2010), and treatment dosage (Meinzer & Breitenstein, 2008). These and other studies may ultimately shed light on the types of aphasia that are likely amenable to remediation with a given treatment approach and to whether damage to specific areas of the brain and/or specific functional activation changes can help to predict response to treatment for anomia (e.g., Fridriksson, 2010; Marcotte & Ansaldo, 2010). To conclude, the overall results further support the efficacy of PCA treatment for the remediation of naming deficits in aphasia. In addition, the neuroimaging data suggest the possibility that active engagement of an individual in his/her therapy (in this case choosing phonological attributes of a target word) may exercise executive functions important for success in treating anomia. The results also underscore the fact that continued exploration of task factors that may promote even better treatment effects using this protocol in individuals with aphasia is warranted, as is continued investigation of the neural underpinnings associated with these treatment effects. Acknowledgements Portions of this work were presented at the Festival of International Conferences on Caregiving, Disability, Aging and Technology, Toronto, Canada, June 2011 (Rochon, E., Leonard, C., Laird, L., Burianova, H., Soros, P., Graham, S., Grady, C.); at The Heart and Stroke Foundation Centre for Stroke Recovery 9th Annual Scientific Meeting, Toronto, Canada, May 2011 (Rochon, E., Leonard, C., Laird, L., Simic, T.); at the Venice Summer School on Aphasia Rehabilitation, Venice, Italy, September, 2011; and at the 14th International Aphasia Rehabilitation Conference, Montreal, Canada, June 2010 (Leonard, C., Rochon, E., Laird, L). We acknowledge the support of Toronto Rehabilitation Institute, which receives funding under the Provincial Rehabilitation Research Program from the Ministry of Health and Long-Term Care in Ontario. We thank the MRI technologists and Caron Murray at Sunnybrook Health Sciences Centre for assistance with this experiment. Danna Rybko and Liz Pigott provided valuable assistance on this project. We especially thank the participants for their patience and perseverance, and we thank the referring clinicians at the Aphasia Institute and the York-Durham Aphasia Centre.

Funding This research was supported by the Heart and Stroke Foundation of Canada [grant numbers 5379 and 6092].

Disclosure statement No potential conflict of interest was reported by the authors.

18

C. Leonard et al.

Notes 1. 2. 3.

P2 had only 21 target words as he did not have enough words in error at baseline. The responses used to represent the rhyming words were identified in a norming study of 10 healthy adults as the one that rhymed best. We did not compare performance on treated versus untreated items because of a confound related to the untreated items. Many of the items in the untreated set were ones that the patient was able to name correctly pre-treatment, resulting in artificially inflated pre-treatment scores for untreated items.


References Beeson, P. M., & Robey, R. R. (2006). Evaluating single-subject treatment research: Lessons learned from the aphasia literature. Neuropsychology Review, 16, 161–169. doi:10.1007/s11065-0069013-7 Busk, P. L., & Serlin, R. C. (1992). Meta-analysis for single-case research. In T. R. Kratochwill & J. R. Levin (Eds.), Single-case research design and analysis (pp. 187–212). Hillsdale, NJ: Laurence Erlbaum Associates. Calvert, G. A., Brammer, M. J., Morris, R. G., Williams, S. C. R., King, N., & Matthews, P. M. (2000). Using fMRI to study recovery from acquired dysphasia. Brain and Language, 71, 391–399. doi:10.1006/brln.1999.2272 Cappa, S. F. (2011). The neural basis of aphasia rehabilitation: Evidence from neuroimaging and neurostimulation. Neuropsychological Rehabilitation, 21, 742–754. doi:10.1080/ 09602011.2011.614724 Coelho, C. A., McHugh, R. E., & Boyle, M. (2000). Semantic feature analysis as a treatment for aphasic dysnomia: A replication. Aphasiology, 14, 133–142. doi:10.1080/026870300401513 Cornelissen, K., Laine, M., Tarkiainen, A., Järvensivu, T., Martin, N., & Salmelin, R. (2003). Adult brain plasticity elicited by anomia treatment. Journal of Cognitive Neuroscience, 15, 444–461. doi:10.1162/089892903321593153 Cox, R. W. (1996). AFNI: Software for analysis and visualization of functional magnetic resonance neuroimages. Computers and Biomedical Research, 29, 162–173. doi:10.1006/cbmr.1996.0014 DeLeon, J., Gottesman, R. F., Kleinman, J. T., Newhart, M., Davis, C., Heidler-Gary, J., … Hillis, A. E. (2007). Neural regions essential for distinct cognitive processes underlying picture naming. Brain, 130, 1408–1422. doi:10.1093/brain/awm011 Efron, B., & Tibshirani, R. (1986). Bootstrap methods for standard errors, confidence intervals and other measures of statistical accuracy. Statistical Science, 1, 54–75. doi:10.1214/ss/1177013815 Fillingham, J., Sage, K., & Lambon Ralph, M. (2005a). Further explorations and an overview of errorless and errorful therapy for aphasic word-finding difficulties: The number of naming attempts during therapy affects outcome. Aphasiology, 19, 597–614. doi:10.1080/ 02687030544000272 Fillingham, J. K., Sage, K., & Lambon Ralph, M. A. (2005b). Treatment of anomia using errorless versus errorful learning: Are frontal executive skills and feedback important? International Journal of Language and Communication Disorders, 40, 505–523. doi:10.1080/ 13682820500138572 Fillingham, J. K., Sage, K., & Lambon Ralph, M. A. (2006). The treatment of anomia using errorless learning. Neuropsychological Rehabilitation, 16, 129–154. doi:10.1080/ 09602010443000254 Foygel, D. (2000). Models of impaired lexical access in speech production. Journal of Memory and Language, 43, 182–216. doi:10.1006/jmla.2000.2716 Francis, D. R., Clark, N., & Humphreys, G. W. (2002). Circumlocution-induced naming (CIN): A treatment for effecting generalisation in anomia? Aphasiology, 16, 243–259. doi:10.1080/ 02687040143000564 Fridriksson, J. (2010). Preservation and modulation of specific left hemisphere regions is vital for treated recovery from anomia in stroke. The Journal of Neuroscience, 30, 11558–11564. doi:10.1523/JNEUROSCI.2227-10.2010 Fridriksson, J., Bonilha, L., Baker, J. M., Moser, D., & Rorden, C. (2010). Activity in preserved left hemisphere regions predicts anomia severity in aphasia. Cerebral Cortex, 20, 1013–1019. doi:10.1093/cercor/bhp160


Aphasiology

19

Fridriksson, J., Richardson, J. D., Baker, J. M., & Rorden, C. (2011). Transcranial direct current stimulation improves naming reaction time in fluent aphasia: A double-blind, sham-controlled study. Stroke, 42, 819–821. doi:10.1161/STROKEAHA.110.600288 Goodglass, H., Kaplan, E., & Barresi, B. (2001). Boston diagnostic aphasia examination (BDAE). Baltimore, MD: Lippincott Williams & Wilkins. Hickin, J., Best, W., Herbert, R., Howard, D., & Osborne, F. (2002). Phonological therapy for wordfinding difficulties: A re-evaluation. Aphasiology, 16, 981–999. doi:10.1080/ 02687030244000509 Howard, D., & Patterson, K. (1992). The pyramids and palm trees test. Suffolk: Thames Valley Test. Huber, H., Poek, K., Weniger, D., & Willmes, K. (1983). Aachener aphasie test. Göttingen: Hogrefe. Kay, J., Lesser, R., & Coltheart, M. (1992). Psycholinguistic assessments of language processing in aphasia (PALPA). East Sussex: Erlbaum. Kochiyama, T., Morita, T., Okada, T., Yonekura, Y., Matsumura, M., & Sadato, N. (2005). Removing the effects of task-related motion using independent-component analysis. NeuroImage, 25, 802–814. doi:10.1016/j.neuroimage.2004.12.027 Lambon Ralph, M. A., Snell, C., Fillingham, J. K., Conroy, P., & Sage, K. (2010). Predicting the outcome of anomia therapy for people with aphasia post CVA: Both language and cognitive status are key predictors. Neuropsychological Rehabilitation, 20, 289–305. doi:10.1080/ 09602010903237875 Leonard, C., Rochon, E., & Laird, L. (2008). Treating naming impairments in aphasia: Findings from a phonological components analysis treatment. Aphasiology, 22, 923–947. doi:10.1080/ 02687030701831474 Leśniak, M., Bak, T., Czepiel, W., Seniów, J., & Członkowska, A. (2008). Frequency and prognostic value of cognitive disorders in stroke patients. Dementia and Geriatric Cognitive Disorders, 26, 356–363. doi:10.1159/000162262 Lurito, J. T., Kareken, D. A., Lowe, M. J., Chen, S. H. A., & Mathews, V. P. (2000). Comparison of rhyming and word generation with FMRI. Human Brain Mapping, 10, 99–106. doi:10.1002/ 1097-0193(200007)10:33.0.CO;2-Q Marcotte, K., & Ansaldo, A. I. (2010). The neural correlates of semantic feature analysis in chronic aphasia: Discordant patterns according to the etiology. Seminars in Speech and Language, 31, 52–63. doi:10.1055/s-0029-1244953 McIntosh, A. R., Bookstein, F. L., Haxby, J. V., & Grady, C. L. (1996). Spatial pattern analysis of functional brain images using partial least squares. NeuroImage, 3, 143–157. doi:10.1006/ nimg.1996.0016 McIntosh, A. R., & Lobaugh, N. J. (2004). Partial least squares analysis of neuroimaging data: Applications and advances. NeuroImage, 23, S250–S263. doi:10.1016/j. neuroimage.2004.07.020 McNeil, M. R., Small, S. L., Masterson, R. J., & Fossett, T. R. D. (1995). Behavioral and pharmacological treatment of lexical-semantic deficits in a single patient with primary progressive aphasia. American Journal of Speech Language Pathology, 4, 76–87. doi:10.1044/10580360.0404.76 Meinzer, M., & Breitenstein, C. (2008). Functional imaging studies of treatment-induced recovery in chronic aphasia. Aphasiology, 22, 1251–1268. doi:10.1080/02687030802367998 Meinzer, M., Flaisch, T., Breitenstein, C., Wienbruch, C., Elbert, T., & Rockstroh, B. (2008). Functional re-recruitment of dysfunctional brain areas predicts language recovery in chronic aphasia. NeuroImage, 39, 2038–2046. doi:10.1016/j.neuroimage.2007.10.008 Meinzer, M., Harnish, S., Conway, T., & Crosson, B. (2011). Recent developments in functional and structural imaging of aphasia recovery after stroke. Aphasiology, 25, 271–290. doi:10.1080/ 02687038.2010.530672 Menke, R., Meinzer, M., Kugel, H., Deppe, M., Baumgärtner, A., Schiffbauer, H., … Breitenstein, C. (2009). Imaging short- and long-term training success in chronic aphasia. BMC Neuroscience, 10, 118. doi:10.1186/1471-2202-10-118 Perani, D., Cappa, S. F., Tettamanti, M., Rosa, M., Scifo, P., Miozzo, A., … Fazio, F. (2003). A fMRI study of word retrieval in aphasia. Brain and Language, 85, 357–368. doi:10.1016/ S0093-934X(02)00561-8 Postman-Caucheteux, W. A., Birn, R. M., Pursley, R. H., Butman, J. A., Solomon, J. M., Picchioni, D., … Braun, A. R. (2010). Single-trial fMRI shows contralesional activity linked to overt


20

C. Leonard et al.

naming errors in chronic aphasic patients. Journal of Cognitive Neuroscience, 22, 1299–1318. doi:10.1162/jocn.2009.21261 Riddoch, M. J., & Humphreys, G. W. (1993). Birmingham object recognition battery. Hove: Lawrence Erlbaum Associates. Roach, A., Schwartz, M. F., Martin, N., Grewal, R. S., & Brcher, A. (1996). The Philadelphia naming test: Scoring and rationale. Clinical Aphasiology, 24, 121–133. Robertson, I. H., & Murre, J. M. J. (1999). Rehabilitation of brain damage: Brain plasticity and principles of guided recovery. Psychological Bulletin, 125, 544–575. doi:10.1037/00332909.125.5.544 Robey, R. R., & Beeson, P. M. (2005, November 18). Aphasia treatment: Examining the evidence. Paper presented at the American Speech-Language Hearing Association Annual Convention, San Diego, CA. Rochon, E., Leonard, C., Burianova, H., Laird, L., Soros, P., Graham, S., & Grady, C. (2010). Neural changes after phonological treatment for anomia: An fMRI study. Brain and Language, 114, 164–179. doi:10.1016/j.bandl.2010.05.005 Sebastian, R., & Kiran, S. (2011). Task-modulated neural activation patterns in chronic stroke patients with aphasia. Aphasiology, 25, 927–951. doi:10.1080/02687038.2011.557436 Sohlberg, M. M., & Mateer, C. A. (2001). Cognitive rehabilitation: An integrative neuropsychological approach. New York, NY: Guilford Press. Spencer, K. A., Doyle, P. J., McNeil, M. R., Wambaugh, J. L., Park, G., & Carroll, B. (2000). Examining the facilitative effects of rhyme in a patient with output lexicon damage. Aphasiology, 14, 567–584. doi:10.1080/026870300401324 Turner, G. R., & Levine, B. (2004). Disorders of executive functioning and self awareness. In J. Ponsford (Ed.), Cognitive and behavioral rehabilitation: From neurobiology to clinical practice (pp. 224–268). New York, NY: Guilford Press. Van Hees, S., Angwin, A., McMahon, K., & Copland, D. (2013). A comparison of semantic feature analysis and phonological components analysis for the treatment of naming impairments in aphasia. Neuropsychological Rehabilitation, 23, 102–132. doi:10.1080/09602011.2012.726201 Vitali, P., Tettamanti, M., Abutalebi, J., Ansaldo, A.-I., Perani, D., Cappa, S. F., & Joanette, Y. (2010). Generalization of the effects of phonological training for anomia using structural equation modelling: A multiple single-case study. Neurocase, 16, 93–105. doi:10.1080/ 13554790903329117 Wiegel-Crump, C., & Koenigsknecht, R. A. (1973). Tapping the lexical store of the adult aphasic: Analysis of the improvement made in word retrieval skills. Cortex, 9, 411–418. doi:10.1016/ S0010-9452(73)80040-1 Yeung, O., Law, S.-P., & Yau, M. (2009). Treatment generalization and executive control processes: Preliminary data from Chinese anomic individuals short report. International Journal of Language and Communication Disorders, 44, 784–794. doi:10.1080/13682820902929081