Search

In this study, two different communicative genres (explanation vs. rport) were elicited in 38German preschool children at the age of 4 years. In one part of the study, explanations of agame were elicited from the child. The game involved spatial movements and figures withvarious geometrical shapes. In a subsequent part, children reported about a puppet and its oddbehaviour to their caregiver. We examined childrens viewpoint in iconic co-speech gesturesand related it to the childrens event structures and linguistic structures that differed in termsof transitivity. Our findings suggest that children do not use viewpoints in a unified waywhich had been reported from studies with adults. In contrast, our results indicate a greatvariability in the ways children use viewpoint in iconic co-speech gesture. We found thatdifferent communicative genres (explanation vs. report) evoke different viewpoints in gesture,due to their different event structure and linguistic structure. During the genre “explanation”,O-VPT gestures occurred more frequently with intransitive utterances, whereas during thegenre “report”, C-VPT gestures occurred more frequently with transitive utterances. Moreover,neither of the events within the communicative genres exclusively evoked one specificviewpoint. Ulrich Mertens, Olga Abramov, Anne Németh, Friederike Kern, Stefan Kopp, and Katharina J. Rohlfing

Previous work by [1] studied gesture-speech interaction in adults. [1] focussed on temporal and semantic coordination of gesture and speech and found that while adult speech is mostly coordinated (or redundant) with gestures, semantic coordination increases the temporal synchrony. These observations do not necessarily hold for children (in particular with respect to iconic gestures, see [2]), where the speech and gesture systems are still under development. We studied the semantic and temporal coordination of speech and gesture in 4-year old children using a corpus of 40 children producing action descriptions in task oriented dialogues. In particular, we examined what kinds of information are transmitted verbally vs. non-verbally and how they are related. To account for this, we extended the semantic features (SFs) developed in [3] for object descriptions in order to include the semantics of actions. We coded the SFs on the children’s speech and gestures separately using video data. In our presentation, we will focus on the quantitative distribution of SFs across gesture and speech. Our results indicate that speech and gestures of 4-year olds are less integrated than those of the adults, although there is a large variability among the children. We will discuss the results with respect to the cognitive processes (e.g., visual memory, language) underlying children’s abilities at this stage of development. Our work paves the way for the cognitive architecture of speech-gesture interaction in preschoolers which to our knowledge is missing so far.

People use speech and gesture together when describing an event or action, where both modalities have different expressive opportunities (Kendon, 2004). One question is how the two modalities are semantically coordinated, i.e. how meaning is distributed across speech and accompanying gestures. While this has been studied only for adult speakers so far, in this paper we present a study on how young children (4 years of age) semantically coordinate speech and gesture, and how this relates to their cognitive and (indirectly) their verbal skills. Results indicate significant positive correlations between cognitive skills of the children and gesture-speech coordination. In addition, high cognitive skills correlate with the number of semantically relevant child descriptions revealing a link between verbal and cognitive skills.