On the Virtues of Assuming Minimal Differences in Information Processing in Individuals and Groups

The metaphor of groups as information processors has been gaining in popularity. If this metaphor’s utility is to be fully exploited, it is first necessary to distinguish - conceptually, theoretically, and empirically - when and how information processing in groups is any different from information processing in individuals. In this paper, several varieties or levels of group information processing are first conceptually distinguished. It is argued that differences between individual and group information processing which are manifest in differences between individual and group output are of greatest interest. However, it is further argued that such individual-group differences in output are not unfailing indicators of such interesting differences in processing. Two empirical illustrations are provided. It is further argued that (a) predictive models assuming little or no individual-group difference in information processing are particularly useful for detecting genuine instances of distinctive group information processing, and (b) that social combination models (such as Davis’ social decision scheme model, 1973) provide one particularly useful class of such models.