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

Locally coherent phrases drive delta band phase synchronization

Poster D50 in Poster Session D with Social Hour, Friday, October 7, 5:30 - 7:15 pm EDT, Millennium Hall

M. Blake Rafferty1, Tim Saltuklaroglu1, Kevin Reilly1, Eun Jin Paek1, Devin M. Casenhiser1; 1University of Tennessee Health Science Center

Recent findings in electrophysiology have shown that low-frequency neural oscillations may become synchronized at the boundaries of constituent phrases, possibly indexing the internal generation of hierarchical syntactic structures. However, psycholinguistic evidence also suggests that such constituent boundaries are often underdetermined or ambiguous during online processing, due in part to the fact that language is produced and perceived through serial channels. In the present study, we investigated the possibility that low-frequency oscillations may instead be driven into synchrony at the boundaries of locally coherent phrases as they are incrementally constructed online, rather than constituent boundaries within global (i.e., sentence-level) syntactic structure. To do this, we recorded EEG for 23 participants (17 female and 6 male) while they read syntactically well-formed English or jabberwocky sentences (e.g., The boy kicked the ball in the net or The pob hinked the doal in the jarm) and ungrammatical English or jabberwocky word-lists, one word at a time. To assess cortical synchronization to syntactic units, we calculated delta band (1-4Hz) inter-trial phase (ITPC) coherence at each word. In line with previous work, we anticipated increased synchrony at the boundaries of syntactic units, with the expectation that these may correspond to sentence-level constituent boundaries (e.g., boy, ball, and net in the English sentence above). However, given that words were presented one at a time, we anticipated that verbs (e.g., kicked) or prepositions (e.g., in)—which, do not comprise phrase boundaries within the sentence-level hierarchical structure—may also elicit increased ITPC, since each can comprise phrase boundaries in locally licensed sentence configurations (e.g., The boy kicked and The boy kicked the ball in are both well-formed English sentences). Monte-Carlo permutation tests with false-discovery rate correction revealed that English and jabberwocky sentences, both of which contained well-formed syntax, elicited significantly greater delta ITPC in comparison to the wordlists. These differences were observed following all items that could form phrase boundaries in their local contexts, even if they did not correspond to constituent boundaries within the global hierarchical structure. These findings provide neurophysiological evidence that syntactic structures are likely generated in incremental fashion, based on local syntactic coherency.

Topic Areas: Syntax, Reading

SNL Account Login

Forgot Password?
Create an Account