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

Poster D10, Thursday, November 9, 6:15 – 7:30 pm, Harborview and Loch Raven Ballrooms

A syntax area in the posterior superior temporal sulcus

William Matchin1, Gregory Hickok2;1UC San Diego, 2UC Irvine

Prominent hypotheses about the neural implementation of syntax ascribe a central role for the left inferior frontal gyrus (LIFG), typically the pars opercularis or pars triangularis, in the generation of basic syntactic structure (Hagoort, 2005; Friederici, 2011; Nelson et al., 2017). However, the hypothesis of a central role for the LIFG in syntax is untenable due to the fact that lesions to this region do not impair basic sentence comprehension or production (Mohr et al., 1978) or acceptability judgments (Linebarger et al., 1983), basic expectations for a brain region involved in core structural aspects of language. The prominent alternative hypothesis for the neural locus syntactic structure building is the anterior temporal lobe (ATL), as this region’s activity reliably correlates with the presence of linguistic structure (Mazoyer et al., 1993; Rogalsky & Hickok, 2009; Pallier et al., 2011; Brennan & Pylkkanen, 2016). However, the ATL is an equally unlikely candidate for the localization of syntax as damage to this region does not result in syntactic deficits in speech production or comprehension (Garrard & Hodges, 2000) and does not impair sentence comprehension beyond the lexical level (Mesulam, 2015). We present a new framework for understanding the neural basis of syntax by positing that an area in the posterior temporal lobe in the ventral bank of the superior temporal sulcus is the locus of hierarchical syntax. We call this region the pSTS syntax area. Previous research has associated this region with lexical processing rather than combinatorial processing (see Hickok & Poeppel, 2007 and Lau et al., 2008 for reviews). However, most researchers in syntactic theory, psycholinguistics, and computational parsing assume at least some “lexicalized” syntax (e.g. Vosse & Kempen, 2000; Joshi et al., 1975; Lewis & Vasishth, 2005; Demberg & Keller, 2008), i.e. stored structure, suggesting a basic overlap in lexical and syntactic processing. Therefore, we posit that the pSTS syntax area houses lexicalized syntactic representations. We review neuroimaging (fMRI, MEG) and aphasia evidence that converges on the pSTS as a core area for lexical-syntactic processing. The pSTS syntax area shows a neuroimaging profile consistent with activation of syntactic representations used for sentence processing in a predictive fashion (Matchin et al., 2017; Matchin et al., submitted). Damage to this region impairs sentence comprehension above and beyond word comprehension (Pillay et al., 2017; Rogalsky et al., submitted), and impairs syntactic acceptability judgments (Wilson & Saygin, 2004). The pSTS syntax area is localized directly in between brain regions specialized for processing speech, orthography, and visual motion. It is also localized directly in between brain regions that are involved in processing semantic information in the ATL and angular gyrus (Binder et al., 2009). Thus, the pSTS syntax area is ideal for relating language data from the external world to semantic representations, in accordance with the classic view that language is the connection between sound and meaning (de Saussure, 1916; Chomsky, 1965).

Topic Area: Grammar: Syntax

Back to Poster Schedule