The relation between the encoding of motion event information and viewpoint in English-accompanying gestures

Properties of motion events (e.g., path, manner) and point of view (e.g., character’s point of view, observer’s point of view) can both be encoded in co-speech gestures. How are these two dimensions of meaning associated? In an examination of English narrative data collected under controlled circumstances, we found that gestures encoding manner of motion were significantly more likely to appear in character viewpoint. Gestures encoding path (but not manner), on the other hand, were significantly more likely to occur in observer viewpoint. Gestures encoding both path and manner simultaneously were also significantly more likely to occur in observer viewpoint. We suggest that selection of point of view may have effects on the encoding of certain semantic features in gesture.