Abstract
Morphosyntactic information does not seem to modulate parafoveal processing in linear inflectional morphology languages (English (Kambe, 2004), Finnish (Bertram, 2003), Malay (Winskel et al., 2014)). To adjudicate between serial and parallel accounts (EZ-Reader (Reichle et al., 2009) vs SWIFT (Engbert et al., 2005)), such studies report only early first-pass measures: first-fixation (FF), single-fixation (SF) and gaze-duration (GD). Stoops et al., (2017; 2019) found early and late morphological effects on words n+1 and n, using boundary-change techniques (Rayner, 1975) in Russian, another linear-morphology language. The preview cost for morphologically-related versus identical and nonword previews was observed in total time (TT) for word n+1 and across early and late measures on pre- and post-boundary and whole-word regions in the word n, supporting the idea of increased attention span for word n (Juhasz et al., 2009). Importantly, the syntactic context of the target word was relatively unrestricted: morphologically-related preview was the most expected syntactic continuation according to a Cloze test (78%). The present study examined if the syntactic predictability modulated parafoveal processing of morphosyntactic information in Russian on the word n and word n+1 by reversing syntactic expectations for the preview. In the two experiments, the identical preview has the expected case marker (94% Cloze test score) and the morphologically-related preview has a syntactically unacceptable case inflection – second subject. Morphologically-related preview on word n+1 didn’t differ from identical and induced preview benefit versus nonword in TT, suggesting word-level facilitation but no early message-level morphological preprocessing. Word n manipulation yielded preview benefit for identical and related versus nonword previews at post-boundary SF and preview cost versus identical but no difference from nonword at whole-word TT. Results support word-level facilitation without message-level integration of syntactically unacceptable morphology. More cross-linguistic investigations are needed to understand the role of syntactic predictability on parafoveal processing in reading.