Presentation

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

Assessing Parallel Word Processing in Deaf Readers with an ERP Flanker Paradigm

Poster B57 in Poster Session B and Reception, Thursday, October 6, 6:30 - 8:30 pm EDT, Millennium Hall
Also presenting in Poster Slam B, Thursday, October 6, 6:15 - 6:30 pm EDT, Regency Ballroom

Brennan Terhune-Cotter¹, Katherine Midgley², Phillip Holcomb², Karen Emmorey²; ¹San Diego State University / University of California, San Diego, ²San Diego State University

While much of reading occurs serially – words are read one after the other – we can also process lexical information across multiple words at once. Evidence for parallel reading processes in hearing people comes from flanker studies, which observe altered behavioral and ERP responses when related or unrelated words are presented on either side of (i.e., flanking) the target word. Those changes are called parafoveal-on-foveal (PoF) orthographic integration effects, in which orthographic information in the parafovea influences the cognitive processing of words in the fovea (the center of vision). Flanker studies have not yet been used with the deaf population; however, nonlinguistic psychophysical and linguistic eye-tracking experiments have shown that deaf people have a wider perceptual span across the visual field in both reading-based and nonlinguistic tasks. For nonlinguistic tasks, this effect is hypothesized to be a compensatory change due to deafness as visual attentional resources are reallocated toward the periphery to compensate for the loss of peripheral auditory information. Such a reallocation of nonlinguistic visual attentional resources may influence higher-level reading behavior by enhancing the perceptual influence of upcoming words that appear in the parafovea. In this ERP study, we presented a lexical decision task in which the centrally presented target word (or pseudoword) was sometimes accompanied by flanker words (or pseudowords) which were either identical to or different from the target item. We had two goals: (1) to replicate results from a previous ERP study using the lexical flanker paradigm (Snell et al., 2019), and (2) to examine how these results extend to the deaf population. Deaf ASL signers were compared with hearing non-signers matched on age and reading skill. In an early (50-150ms) window, larger negativities were observed for both flanker conditions compared to the no flanker condition in both groups, likely reflecting the added complexity of the flanked arrays (three words rather than one). In line with Snell et al. (2019), we found differences between the identical-flanker and different-flanker condition beginning after ~200ms and continuing past 600ms, with a larger N250 and N400 when flankers were different from the target. These findings support the hypotheses that the N250 and N400 respectively reflect abstract sublexical and lexicosemantic processing independent of the relative word location in the visual field. The N250 was even greater for different- vs. no-flanker conditions, as in Snell et al. (2019), indicating greater sublexical competition arising from the presence of flankers regardless of flanker identity. The presence of PoF effects after 200ms supports the parallel reading hypothesis at the sublexical (bigram) and lexicosemantic levels of word processing. Both deaf and hearing readers showed clear PoF effects, but the N250 for different vs. identical flankers emerged earlier for deaf readers, which might indicate more fine-grained orthographic processing of parafoveal words for deaf readers, as opposed to processing at the bigram and whole-word levels for hearing readers. (Snell, J., Meade, G., Meeter, M., Holcomb, P., & Grainger, J. (2019). An electrophysiological investigation of orthographic spatial integration in reading. Neuropsychologia, 129, 276–283.)

Topic Areas: Signed Language and Gesture, Reading