| Abstract: |
Background: Renewed interest in the effects of intensity on treatment has led to development of short-term, intensive treatment protocols, such as Constraint-Induced Language Therapy (CILT), in which participants with chronic aphasia begin to show statistically significant language improvements in as little as 2 weeks. Given its relatively short treatment cycle, CILT is also a good choice of treatment methodology for studying brain/behaviour plasticity in post-stroke aphasia. Aims: This study aimed to examine differences between two short, intensive treatment protocols in a participant with chronic Wernicke's aphasia both in terms of treatment outcomes and changes in patterns of BOLD signal activation. Methods and Procedures: The participant (ACL) participated in language testing and an fMRI overt speech confrontation-naming paradigm pre and post 2 weeks of CILT, post 2 weeks of unconstrained language therapy (PACE), and 6 months post-CILT. He named 48 black/white line drawings from each of four conditions: treated (CILT or PACE), untreated, or consistently correctly named pictures. Outcomes and Results: Naming treated pictures improved, even in the scanner, while naming untreated pictures did not. About one third of PACE and three-fourths of CILT gains were maintained. Rather than a distinct pattern of activation distinguishing treated from untreated or CILT from PACE pictures, ACL recruited a frontal network during naming of all pictures that included left middle and inferior frontal cortex, SMA and pre-SMA, and that varied in spatial extent and degree of activation according to accuracy and performance expectation. In post-hoc analyses of accuracy, this frontal network was most active during incorrect trials. At 6 months post-CILT, compared to controls, incorrect naming recruited a large and significant bilateral network including right Wernicke's area homologue. Conclusions: Results suggest that short, intensive therapy can improve naming and jumpstart language recovery in chronic aphasia, whether responses are constrained to the speech modality or not. Modulation of a left frontal network was associated with accuracy in naming and may represent compensatory adaptation to improve response selection, self-monitoring, and/or inhibition. |
