[PS-2.87] Visual constraints on the stereotypical predictability of agents during situated language comprehension

Rodríguez, A. 1 , Burigo, M. 1 & Knoeferle, P. 2

1 Cognitive Interaction Technology Excellence Cluster, Bielefeld University, Bielefeld
2 Department of German Language and Linguistics, Humboldt University, Berlin

We investigated how constraints on the concurrent visual context modulate the use of prior gender and action cues and stereotypical knowledge during situated language comprehension. Participants saw videos of hands performing actions over objects and then inspected the pictures of potential agents' (one female one male) while verifying the match between prior events and auditory German OVS sentences. Unlike in our previous experiments (Rodríguez et al. 2015), the concurrent context also showed a picture of the videotaped object and a ?competitor object?, which could be mentioned in the sentence. Fixations to the agents' faces were measured during comprehension. We manipulated the match between prior actions and the actions described by the sentence and the match between the stereotypical valence of the described actions and the gender of the videotaped agent. We replicated the visual preference for the target agent's face (whose gender matched the videotaped hands) from Rodríguez et al., reduced by action-sentence mismatches. However, this time effects emerged earlier (object region). Additionally, stereotypicality modulated these effects. Results suggest that the visual availability of objects during sentence comprehension facilitates the activation of representations from recent events (speeding up mismatch effects) as well as of alternative unseen events, favoring stereotypical expectations.