PROJECT ABSTRACT Aphasia affects approximately one third of stroke survivors (Berthier, 2005), and between 30 and 43% of these remain severely affected in the long term (e.g. (Bakheit et al., 2007). With a rising burden of stroke, it is increasingly important to develop effective and individualized treatment for patients with aphasia. An understanding of the brain mechanisms that underlie plasticity and recovery is therefore essential. However, the problem to solve is far from trivial given that language impairment and recovery after stroke is impacted by many factors including premorbid differences in brain organization, size and site of lesion, and post-stroke behaviors. In the proposed work, we examine the brain regions that are capable of subserving language recovery. While the prevailing hypothesis is that language recovery is mediated by the regions of the fronto-temporal network – either the ipsilesional left-hemisphere (LH) regions or the homologous right-hemisphere (RH) regions, we examine a novel hypothesis that a domain-general fronto-parietal multiple demand (MD) network may also mediate language recovery in patients with aphasia (PWA). We propose to evaluate three possible changes in the functional architecture of the MD and language networks: (i) response magnitude in each network during language processing; (ii) intra-network synchronization among MD regions, and among language regions; and (iii) across-network synchronization between MD and language regions. We will first characterize post-stroke differences in each measure by comparing 40 chronic PWA and 40 age-matched controls. Then, we will identify individual differences in neural signatures that co-vary with differences in language performance on standard behavioral tests. Finally, we will assess the causal contribution of the MD network to language recovery in a subset of 40 PWA tested at two time points, both in fMRI and behavioral tasks. This work is innovative because it uses a suite of methodological tools to facilitate localize language processing in PWA by accounting for variability in lesion site, volume and anatomy-function mapping. Further, we examine a subset of patients at two time points, allowing a within group comparison of changes in the MD network over time and as a function of recovery. A key outcome of this project is the development of a comprehensive explanation for mechanisms of neuroplasticity and reorganization of language function in aphasia. Another outcome of the project is the delineation of neural phenotypes – including in the acute stage – that explain the variability in linguistic and cognitive performance of PWA while accounting for variability in anatomy-function mapping. A third important outcome will be to pave the way for new behavioral and neuro-stimulation therapies that capitalize on the engagement of the MD network in the service of language recovery. Establishing the potential recruitment of non-language regions (i.e., multiple demand ...