Rapid visual form-based processing of (some) grammatical features in parallel reading: An EEG study in English

Abstract

AbstractTheories of language processing – and typical experimental methodologies – emphasize word-by-word processing of sentences. This paradigm is good for approximating speech or careful text reading, but arguably, not for the common, cursory glances used while reading short sentences (e.g., cellphone notifications, social media posts). How much grammatical information can be gleaned from a single glance? In an electroencephalography (EEG) study, brain responses to grammatical (the dogs chase a ball) stimuli diverged from scrambled counterparts (a dogs chase ball the) ∼300ms post-sentence onset, and from non-lexical consonant strings (thj rjxb zkhtb w lhct) ∼220ms post-sentence onset. This demonstrates early recognition and cursory analysis of linguistic stimuli. However, EEG responses do not diverge between grammatical sentences and their counterparts with ungrammatical agreement (the dogs chases a ball). Additionally, the surprisal of individual words affects the EEG signal at non-uniform time points, from 250ms–600ms. We propose that, in cursory reading in a single glance, readers extract some sentence-level information, such as basic syntactic structure, then ‘fill in’ some lexical details in a top-down fashion afterwards. This cursory syntactic analysis, however, is not detailed enough to support detection of formal syntactic agreement errors. We suggest this may be due to either the minimal visual salience of agreement morphology in English (-s), or a strategic ignoring of semantically-inert syntactic features for the sake of extracting a semantic ‘gist.’