The communicative value of conversational hand gestures

Three experiments investigated the extent to which spontaneous gestural accompaniments to a spoken message enhance the message's communicative effectiveness. All three employed a modified referential communication task in which subjects (speakers) were videotaped as they described a stimulus to a partner, who then tried to select it from a set of similar stimuli. Half of the dyads communicated face-to-face; the remainder were in different rooms and communicated over an intercom. The videotaped descriptions were presented to new subjects (listeners), who tried to select the stimulus described. Half of these listeners both saw and heard the videotape; the remainder only heard the soundtrack. The three experiments differed in the type of stimulus the speakers described: abstract graphic designs, novel synthesized sounds, or samples of tea. Communication accuracy (i.e., the rate at which listeners selected the correct stimulus) was better than chance in all three experiments, but in none was accuracy enhanced by allowing the listener to see the speaker's gestures. The results bring into question the assumption that the communication of semantic information is a primary function of conversational hand gestures. (C) 1995 Academic Press, Inc.