Poster A65, Saturday, March 25, 5:00 – 7:00 pm, Pacific Concourse
Visual gender cue effects on incremental language comprehension
Alba Rodriguez1,3, Yoana Vergilova2, Matthew W Crocker2, Pia Knoeferle1; 1Humboldt University Berlin, 2Saarland University, Germany, 3Bielefeld University, Germany
Psycholinguistic studies of gender knowledge effects have often manipulated the match between a linguistic context and words in a subsequent sentence (e.g. finding the pronoun "her" after a sentence talking about a "minister"; see Hammer et al., 2008; Kreiner et al., 2009; Xu, et al., 2013). Gender cues have also been conveyed through pictorial contexts. For instance, event-related brain potential (ERP) studies have shown that mismatches between a picture and the ensuing sentence affect sentence processing in real time (Vissers et al., 2008; Willems et al., 2008; Knoeferle et al., 2011). We examined the effects of prior visual gender cues in videos of male vs. female hands performing an action on participants’ ERPs as they listened to sentences containing a masculine or feminine actor name (translation of German object-verb-subject example sentence: ‘The cake (obj) bakes soon Susanna/Tobias (subj)’). The gender of the hands either matched or mismatched the gender of the actor’s name. At ‘Susanna/Tobias’, mean amplitude N400s were more negative (250-400 ms) for gender mismatches than matches. Additionally, gender mismatches (vs. matches) elicited more positive mean amplitude (500-700 ms). As observed for linguistic contexts, amplitude differences were more pronounced at posterior than anterior sites. The findings suggest that even subtle gender cues – an actor’s hand – elicited semantic expectations for a (gender-matching) actor (name). Failure to confirm these expectations seemed to affect both lexical semantic and integration processes.
Topic Area: LANGUAGE: Semantic