Generality of a Theory of Collective Induction: Face-to-Face and Computer-Mediated Interaction, Amount of Potential Information, and Group versus Member Choice of Evidence

Abstract

In order to assess the generality of a theory of collective induction that has previously been supported for face-to-face interaction (Laughlin and Hollingshead, Organizational Behavior and Human Decision Processes, 1995), three experiments compared face-to-face and computer-mediated interaction for four-person groups for three successive rule induction problems. The predictions of the theory fit the obtained probabilities of group hypotheses for distributions of member hypotheses better than four other plausible models for both face-to-face and computer-mediated groups, over four levels of potential information, and for group versus member choice of evidence. There were comparable numbers of correct hypotheses for face-to-face and computer-mediated interaction, increasing correct hypotheses with increasing potential information, and comparable numbers of correct hypotheses for group versus member choice of evidence. The three experiments indicate the generality of the theory for both traditional face-to-face interaction and computer-mediated interaction, four levels of potential information, and group versus member choice of information. The comparisons of face-to-face and computer-mediated interaction extend the emphasis on social communication processes on decision and idea generation tasks of previous research to social combination processes in the acquisition and processing of new information in cooperative rule learning.