Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
An EEG functional localizer for identifying visual word form responses in sensor and source space
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 D125 in Poster Session D, Wednesday, October 25, 4:45 - 6:30 pm CEST, Espace Vieux-Port
This poster is part of the Sandbox Series.
Dustin A. Chacón1, Donald Dunagan1; 1University of Georgia
Substantial attention has been paid to the left fusiform gyrus' role in processing visual word form information (Tarkiainen, Helenius, Hansen, Cornelissen, & Salmelin 1999; Dehaene & Cohen 2011), and its role in initial stages of morphological processing (Solomyak & Marantz 2011). The functional 'visual word form area' (VWFA) has been studied predominantly in fMRI and MEG, due to their accurate spatial resolution. However, coregistered fMRI-EEG studies on word recognition show that left fusiform activity correlates with the N1 peak in EEG recordings (Cohen, Dehaene, Naccache, Lehéricy, Dehaene-Lambertz, Hénaff, Michel 2000; Pleisch, Karipidis, Brem, Röthlisberger, Roth, Brandeis,Walitza, Brem 2019), and EEG source reconstruction techniques localize N1 responses to bilateral occipitotemporal regions (Brem, Halder, Bucher, Summers, Brandeis 2009; Maurer, Brem, Bucher, Brandeis 2005). Here, we present a replication of an 'abridged' functional localizer, first demonstrated in MEG by Gwilliams, Lewis, & Marantz (2018). We show that this quick (~5 min) experimental paradigm can identify word-specific brain responses in both sensor and source space, which can then serve as a functional ROI. [METHODS] N = 12 English speakers participated in a passive reading task. Stimuli were 50, 4-letter words printed in a sans serif font, and 50, 4 non-linguistic symbols. Both stimuli were embedded in two levels of noise. Luminance, horizontal width, and pixel density of the stimuli were controlled (p > 0.05). EEG signals were recorded with a 64-channel BrainVision actiCHamp+ system, with FCz as on-line reference. Electrode positions were digitized using BrainVision CapTrak system. Stimuli were presented for 60ms, with a 200ms ISI. The study took approximately 5 minutes. [RESULTS] Spatio-temporal cluster-based permutation tests were conducted on sensor space data, from 0–300ms. Sensors were re-referenced off-line to average reference. We identified a cluster 125–168ms, showing greater positive activity for words over symbols, irrespective of noise, over left posterior sensors (p=0.02). Another cluster showed more positive activity for low noise conditions over high noise stimuli, irrespective of language status, over midline posterior sensors (p<0.01). Sensor positions were coregistered with 'fsaverage' template brain, and inverse solutions were computed using sLORETA. Spatio-temporal cluster-based permutation tests were conducted in source space over bilateral occipitotemporal regions, taken from Gwilliams et al. (2018). A marginal cluster showing greater activity for words over symbols was observed in inferior and anterior left temporal lobe, 140–168ms (p=0.06). Expanding the search space to include all left temporal lobe, we found a significant cluster 138–168ms in left anterior middle and superior temporal regions, showing greater activity for words over symbols (p=0.02). Expanding the search space to the entire brain, a cluster was identified in left posterior fusiform gyrus, 137–168ms (p=0.09), showing greater activation for words over symbols. [CONCLUSION] Understanding the brain's response to written words requires a fast, easy-to-use functional localizer. We demonstrate that the 'abridged' functional localizer paradigm used by Gwilliams et al. (2018) in MEG works in EEG. However, our results localize to a more anterior and lateral portions of left temporal lobe than previous studies.
Topic Areas: Writing and Spelling, Methods