Abstract
Through a series of experiments, Cohen and Otterbein (1992) found that sentences presented together with pantomimes or with representational gestures were more often recalled than sentences presented without gestures. However, the mechanism underlying this mnemonic effect of a multimodal presentation was not fully specified. It may depend on activation of mental images and/or on increased distinctiveness of gestured utterances. The present study compared memory for sentences accompanied by different kinds of gestures—representational and nonrepresentational—to assess the role of meaning in remembering the verbal material. The results of the first experiment indicated that presentation of nonrepresentational gestures did not increase sentence recall as representational gestures did. These representational gestures also influenced performance in a sentence recognition task. In the second experiment, videotapes were edited to create mismatches and when sentences were presented with incongruent representational gestures, recall was not facilitated. Thus, in this study, memory scores did not depend on the distinctiveness of sentences presented with meaningless or mismatching gestures. These findings are discussed in relation to the literature on memory for actions in experimenter-performed tasks (EPTs).