Overcoming Learned Non-Use in Chronic Aphasia

In this study the investigators are examining the effectiveness of intensive speech therapy in chronic moderate-to-severe stroke-induced aphasia under two conditions - responses "constrained" or unconstrained to speech. Both treatments involve massed practice communicating, using intensive language action therapy 3 hours/day, 5 days/week for two weeks, followed by six months of a home practice program. One treatment stresses spoken responses as the preferred expressive modality during intensive therapy. Before and after treatment, and following the home practice program and a period of no practice, the investigators will administer several tests and discourse samples to examine changes associated with the treatments. Participants will also undergo structural and functional MRI testing at these time points. The investigators will also attempt to quantify the degree to which improvements following intensive language therapy and home practice correlate with changes in Quality of Life measures as perceived by both participants with aphasia and their significant others. It is hypothesized that, whereas both treatments will lead to improvements in naming practiced words and communicating, outcomes will be enhanced for the group randomly assigned to the "constraint" condition. Moreover, performance will be enhanced on words practiced during the home practice program, including those that were not practiced during intensive therapy. Improved naming will correlate with modulation of 'signature' language and attentional networks, whose variability will depend on remaining viable brain structures. Initial severity and site/extent of lesion should predict patients' ability to transfer gains in naming to improvements in discourse.