Neurophysiological Correlates of Grammatical Structure Processing in Russian

Automatic syntax processing – most commonly reflected in ELAN component for (morpho)syntactic violations – is a well-established phenomenon, but its mechanisms are poorly understood. Two alternative hypotheses are offered: (1) syntactic priming (leading to response reduction for connected morphemes) vs. (2) error detection (response increase for violation). To test these accounts, we used a non-attend design with visual distraction and recorded ERPs to spoken Russian pronoun-verb phrases (e.g., ‘she/he/I walked’) that either agreed or mismatched in subject-verb gender marking as well as to the same critical verbs presented in isolation. Three pronouns and ten different verbs spoken in male and female voices were used. Results revealed both early and late syntax-processing indices, including ELAN, LAN and P600 for agreement violation, suggesting a high degree of automaticity in syntactic parsing. The largest ELAN responses were obtained for isolated verbs, followed by incongruent phrases, suggesting that its reduction for well-formed phrases is underpinned by syntactic priming between words/morphemes, potentially mediated by associative links formed during previous experience. Moreover, we also found ELAN and N400 for those first-person phrases where the speaker voice mismatched the grammatical gender (I walked[fem/masc]). Previously, such integration of extralinguistic features into syntactic analysis was only demonstrated in N400 for attended sentences. Overall, our results support rapid automatic syntactic processing that underpinned by the links between related words/morphemes and relying on both linguistic and extralinguistic information in the auditory input.