thoughts to ordinary listeners

The gestures that spontaneously occur in communicative contexts have been shown to offer insight into a child’s thoughts. The information gesture conveys about what is on a child’s mind will, of course, only be accessible to a communication partner if that partner can interpret gesture. Adults were asked to observe a series of children who participated ‘live’ in a set of conservation tasks and gestured spontaneously while performing the tasks. Adults were able to glean substantive information from the children’s gestures, information that was not found anywhere in their speech. ‘Gesture-reading’ did, however, have a cost ‐ if gesture conveyed different information from speech, it hindered the listener’s ability to identify the message in speech. Thus, ordinary listeners can and do extract information from a child’s gestures, even gestures that are unedited and fleeting. Audio-recorders are frequently used to capture children’s responses, a practice that works particularly well if all of the relevant information appears in the speech stream. However, children do more than just talk. They routinely move their hands as they speak ‐ they gesture. Much work has shown that the gestures produced along with speech display information about the speaker’s thoughts (Kendon, 1980; McNeill, 1992), even when the speaker is a child (Evans & Rubin, 1979; Perry, Church & GoldinMeadow, 1988; Crowder & Newman, 1993). 1 Interestingly, the information conveyed in gesture does not always match the information conveyed in the accompanying speech (McNeill, 1992). For example, when asked to explain her responses to a series of conservation questions, one child highlighted the containers’ heights in speech (‘this one’s taller than that one’) while highlighting their widths in gesture (two vertical palms indicating first