A fundamental aspect of our existence is that spoken language is our primary means of communication with others. Unfortunately, the ability to speak, and multiword (i.e., phrase and sentence) production in particular, is impaired following stroke. It is unclear what drives successful recovery of language production following stroke. The proposal describes a research program that uses cognitive behavioral testing, neuroimaging, and statistical modeling to understand how multiword production and the neural structures that support it change during recovery by testing patients within 72 hours of stroke (acute stage) and again at 1, 6, and 12 months after stroke. Previous neuroimaging and neuropsychological studies, including several from our laboratories, have identified two cognitive abilities and associated brain regions that may be critical for multiword production. These abilities are the ability to resolve interference from word competitors during word selection and the ability to store semantic information in working memory, subserved by posterior and anterior regions of the left inferior frontal gyrus, respectively. However, conclusions about the role of these abilities have been drawn from data collected from chronic stroke cases, months and years following the onset of stroke. As a result, conclusions about brain-behavior relations are problematic due to significant reorganization of function that occurs during recovery. Because studies of chronic patients typically exclude people whose language disorders have resolved, they underestimate the impact of damage to small regions of the brain on multiword production. Further, the way selection and working memory capacities interact to support multiword production remains to be examined. We will use recent innovations in the analysis of longitudinal data (Latent Change Score models) and structural neuroimaging (sequences and modeling approaches to assess damage and hypoperfusion to gray matter and integrity of white matter tracks) to test the hypothesis that selection and working memory and associated brain regions are critical to the recovery of multiword production.
Aim 1 will test the hypothesis that selection and semantic working memory capacities are necessary for multiword production by assessing patients before reorganization occurs, i.e., at the acute stage, within 72 hours of stroke, which will include those with smaller lesions.
Aim 2 will identify the neural structures associated with selection and working memory in the acute stroke patients tested in Aim 1.
Aim 3 will test the hypothesis that recovery of selection and working memory capacities is critical to post stroke language outcome by assessing behavior and neuroanatomical changes across acute, 1, 6, and 12 months post stroke time points. If successful, this research will result in a detailed theory of the cognitive-brain basis of multiword production. This knowledge will have important health implications because information about how language ability is impaired and subsequently recovers is essential for managing the consequences of stroke through the development of language rehabilitation strategies.

Public Health Relevance

This project examines the basis of language recovery in stroke patients. Assessing language abilities while collecting neuroanatomical information over time after stroke will enable us to gain predictive power about what type of brain damage and deficits of cognitive abilities lead to chronic language loss (aphasia), and as a result, we will b able to identify patients likely to require intense rehabilitation therapy soon after their stroke. Moreover, because we will be able to identify neural mechanisms that subserve language recovery, this opens possibilities for targeted clinical therapies. If successful, this research wil provide a clear path to improved diagnosis and treatment of aphasia.

National Institute of Health (NIH)
National Institute on Deafness and Other Communication Disorders (NIDCD)
Research Project (R01)
Project #
Application #
Study Section
Language and Communication Study Section (LCOM)
Program Officer
Cooper, Judith
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Baylor College of Medicine
Schools of Medicine
United States
Zip Code
Hughes, Julie W; Schnur, Tatiana T (2017) Facilitation and interference in naming: A consequence of the same learning process? Cognition 165:61-72
Schnur, Tatiana T (2017) Word selection deficits and multiword speech. Cogn Neuropsychol 34:21-25
Shahid, Hinna; Sebastian, Rajani; Schnur, Tatiana T et al. (2017) Important considerations in lesion-symptom mapping: Illustrations from studies of word comprehension. Hum Brain Mapp 38:2990-3000