Presentation

Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Poster Slams

Functional partitioning of sentence processing and emotional prosody in the right perisylvian cortex of perinatal stroke participants

Poster E44 in Poster Session E, Saturday, October 8, 3:15 - 5:00 pm EDT, Millennium Hall

Kelly Martin¹, Anna Seydell-Greenwald^1,2, William Gaillard^1,3, Peter Turkeltaub^1,2, Elissa Newport^1,2; ¹Center for Brain Plasticity and Recovery, Georgetown University Medical Center, ²MedStar National Rehabilitation Hospital, Washington, DC, ³Children’s National Hospital, Washington, DC

In most adults, ‘core’ language functions, such as sentence processing, are strongly lateralized to the left hemisphere (LH), while other language functions, such as vocal emotion detection (emotional prosody), are oppositely lateralized to the right hemisphere (RH). Indeed, a stroke impacting LH perisylvian regions in adulthood generally produces aphasia but not aprosodia, while a stroke to RH perisylvian regions produces aprosodia but not aphasia. In contrast, a stroke to LH perisylvian regions around the time of birth does not result in chronic aphasia in most cases: studies have shown that language abilities develop in the normal range and language functions that are typically left-lateralized instead recruit homotopic RH perisylvian regions. These findings raise an important question: How does the functional map of right perisylvian regions change to support both emotional prosody and typically left-lateralized language functions after a LH stroke early in development? In the current work we used task fMRI to examine how two processes that lateralize to parallel regions in opposite hemispheres of the healthy brain (sentence processing and emotional prosody processing) are organized in the intact RH of individuals who suffered a large LH stroke around the time of birth. In our recent work we have shown that that sentence processing becomes organized in precisely homotopic RH regions of individual left hemisphere perinatal stroke participants (LHPSPs) without apparent detriment to language or emotional prosody outcomes, and that these two functions appear to claim separate cortical territories in right perisylvian cortex. Here we investigated this finding further by asking about these outcomes in more precise quantitative detail. After a LH perinatal stroke, does sentence processing simply become reflected into RH frontotemporal regions and overlap with emotional prosody processing as one would expect with simple mirroring? Or do these functions overlap less than would be expected if that was the case? Thirteen perinatal stroke participants were included, all of whom suffered a large cortical stroke to the left middle cerebral artery. Eleven healthy controls were also included, who are siblings from the same families and roughly age-matched to the stroke participants. Participants completed an auditory description decision task and an emotional prosody decision task in the scanner. We employed a ‘top voxel’ analysis approach which allows us to compare the same number of the most active voxels between participants and between hemispheres to evaluate similarity in the spatial arrangements of activity. Independent samples t-tests revealed that the frontal and temporal areas recruited for these functions overlapped significantly less in the intact RH of LHPSPs compared to the overlap when LH sentence processing areas in controls were transposed into the RH and spatially compared (using a Dice Coefficient) with their emotional prosody areas. This result indicates that in the intact hemisphere of perinatal stroke participants, separation between the areas that perform sentence processing and the areas that perform emotional prosody processing may be important for these two functions to develop successfully. Our findings have important implications for how the cortical layouts for sentence processing and emotional prosody form during development.

Topic Areas: Development, Disorders: Developmental