Wiggle, Wiggle, Wiggle: How Visual Cues Influence Thematic Role Assignment in Children and Adults

AbstractGerman 5-year-olds are able to rapidly recruit depicted actions to assign thematic roles in unambiguous sentences when these actions can be inspected throughout sentence presentation (Münster, 2016; Zhang & Knoeferle 2012). In two visual-world eye tracking studies, we investigated whether these findings extend to locally structurally ambiguous utterances and to short-lived action presentation. In addition, we compared the action depiction to a character's wiggling motion. The action and the wiggle served as cues to the agent (subject) in difficult-to-understand OVS sentences. Participants listened to structurally ambiguous object-verb-subject (OVS) sentences about, for instance, a bug being pushed by a bull while inspecting a bull, a bug, and a worm. We manipulated the scene at verb-onset such that either a) no action no wiggle, b) no action one wiggle, c) one action no wiggle, or d) one action one wiggle appeared. Both of these animations caused the adults and the children to visually anticipate the agent role filler (corresponding to the subject in the OVS sentence) before its mention. However, in answering post-trial who-does-what-to-whom comprehension questions, the children did not (unlike suggested by previous findings) benefit from the action depictions. Together the eye-gaze and post-trial comprehension results suggest that the nature of cue presentation (e.g., the abrupt onset of an action or a wiggle and limitations on cue presence) plays an important role in both the immediate visual attention and somewhat later interpretation effects of such visual cues during children's language comprehension.

Return to previous page