Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Functional Localizers for American Sign Language Comprehension
There is a Poster PDF for this presentation, but you must be a current member or registered to attend SNL 2023 to view it. Please go to your Account Home page to register.
Poster C109 in Poster Session C, Wednesday, October 25, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
Brennan Terhune-Cotter1, Stephen McCullough1, Karen Emmorey1; 1San Diego State University
When using fMRI to test hypotheses about patterns of neural activation which underlie cognitive processes, researchers commonly constrain activation elicited by experimental tasks with a localizer task, which isolates neural activity unique to that cognitive process. Localizer tasks are commonly designed on a per-experiment basis, so results cannot be directly compared across studies. A solution is to create a standardized localizer task known to reliably activate neural areas of interest, which can then be used to define individual functional regions of interest (fROIs) in each participant. We developed localizer tasks for American Sign Language (ASL) comprehension which are analogous to spoken or written language localizer tasks that have been developed by other research teams (Fedorenko et al., 2010; Scott et al., 2017; Ayyash et al., 2021). Our localizer tasks elicit ASL processing at the lexical, sentential, and narrative levels, with the baseline condition being degraded (blurred) videos of the same stimuli. The lexical condition consists of lists of unrelated nouns or verbs, which also allow us to contrast activation patterns for comprehending each lexical class. The sentential condition uses the same or similar words in simple, thematically unrelated sentences. The narrative condition consists of excerpts from a story (Alice in Wonderland) with narrative and prosodic devices typical of ASL storytelling. We have scanned 15 deaf native or early signers with our localizer tasks. We used the group-constrained subject-specific approach developed in Fedorenko et al. (2010) to define group-level partitions. Here we compare the efficacy of our three localizer tasks in defining partitions which can be used for future studies. The three localizer tasks all activate similar frontotemporal and occipital regions but to different extents. The sentences > baseline contrast elicits the strongest activation, likely due to the added processing demands of comprehending sets of thematically unrelated sentences as compared to individual words or a single familiar story. This contrast produces thirteen meaningful group-level partitions: five partitions over bilateral occipital areas, bilateral fusiform gyri, and left inferior frontal gyrus; seven partitions spanning the bilateral superior and middle temporal gyri; and one partition straddling the left middle frontal and precentral gyri. We used individually defined fROIs generated with this localizer task to assess neural activation associated with processing of syntactic structure (sentences > word lists), lexical items (word lists > baseline), and lexical class (noun vs. verb lists). Both sentential and lexical processing activated bilateral occipital gyri, fusiform gyri, and superior/middle temporal areas; sentential processing also activated left IFG and precentral gyrus. In the direct noun-verb contrasts, nouns did not activate any regions more than verbs, but verbs activated bilateral occipital and fusiform gyri, bilateral temporal areas (with more extensive activation in the left hemisphere), and left IFG and precentral gyrus. Our data indicate that a distributed network involving temporal, fusiform, and occipital regions (the latter possibly indicative of top-down linguistic processes) subserve lexical and sentential processing in ASL. Verbs more heavily engage this network compared to nouns, likely due to their greater semantic and syntactic complexity.
Topic Areas: Phonology and Phonological Working Memory,