You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

 
Poster E20, Friday, November 10, 10:00 – 11:15 am, Harborview and Loch Raven Ballrooms

Functional subspecialization of Broca’s area in the controlled selection of verbal and nonverbal representations and fluent sentence production.

Denise Y. Harvey1,2, Myrna F. Schwartz1;1Moss Rehabilitation Research Institute, 2University of Pennsylvania

The left inferior frontal gyrus (LIFG) has long been implicated in language production, as stroke-related damage to this region often results in an inability to fluently produce sentences, called “Broca’s aphasia”. However, the LIFG has also been shown to play a more general role in controlling the selection of mental representations from among competing alternatives. This suggests that faulty cognitive control underlies impairments in fluent sentence production. Indeed, individuals with Broca’s aphasia, as compared to those with fluent forms of aphasia (hereafter, “NonBroca’s”), perform poorly on tasks that require controlled selection of verbal representations (Schnur et al., 2006). Whether nonverbal representations are similarly affected remains unclear. Thus, the goal of the current study was twofold: (1) to establish the scope of controlled selection impairments in post-stroke aphasia; and (2) assess whether impaired selection is associated with LIFG damage and/or the ability to fluently produce sentences. We collected behavioral and structural neuroimaging data from 18 chronic aphasic speakers. Nine participants presented with Broca’s aphasia, and were contrasted with a group of nine NonBroca’s participants closely matched for performance on measures of verbal and nonverbal processing that minimize control demands. To assess controlled selection abilities, participants named pictures (verbal) and matched associatively related pictures (nonverbal) in blocks of trials depicting objects from the same-semantic category (i.e., high selection demand) versus different categories (i.e., low selection demand). Fluency was assessed using two measures from the Quantitative Production Analysis of narrative speech: proportion of closed-class words and proportion of words in a sentence (relative to all words). We quantified percent damage to the LIFG based on lesion overlap in Brodmann area (BA) 44 and 45. Correlational analyses controlled for the contribution of total lesion volume. Participants were slower to respond in the related compared to the unrelated blocks in both tasks (p < .0001). However, the magnitude of this effect did not differ based on aphasia classification (Broca’s versus NonBroca’s participants; p = .85). Individual estimates of controlled selection ability were correlated with percent damage to a posterior region within the LIFG (BA 44) (r = .49, p < .05). However, controlled selection abilities were not correlated with measures of fluent sentence production (r’s < .2, p’s > .46). Instead, fluent sentence production was correlated with percent damage to a more anterior LIFG region (BA 45) (r’s > -.49, p’s < .05). The findings from this research indicate that the LIFG is involved in both controlled selection over competing mental representations (whether verbal or nonverbal) and fluent sentence production. However, the results suggest a functional subdivision between different regions that comprise of the LIFG: a posterior region involved in controlled selection (BA 44) and an anterior region involved in fluent sentence production (BA 45). Thus, the presumed relationship between LIFG-mediated controlled selection impairments and nonfluent aphasia may instead reflect the co-occurrence of damage to adjacent regions (as constrained by the vascular architecture). Taken together, this research sheds light on how LIFG damage may differentially contribute to language impairments in aphasia.

Topic Area: Language Disorders

Back to Poster Schedule