Face-to-face discussion has been shown to increase cooperative behavio
r in social dilemmas. Two general explanations of this effect were tes
ted: group identity (concern for fellow group members' outcomes) and p
erception of consensus (degree of agreement perceived among group memb
ers on how to respond to the dilemma). Undergraduate students particip
ated in four-person groups in one of four experimental conditions. The
four conditions differed in common fate (interdependent outcome in a
lottery), face-to-face discussion of the dilemma, discussion of an imp
ortant (irrelevant) issue, and no discussion (control). Results indica
te that (a) the common fate manipulation neither aroused group identit
y nor enhanced cooperation, (b) group identity may not be sufficient t
o elicit cooperation in a social dilemma, and (c) the most plausible e
xplanation of group discussion effects is members' perceived consensus
to cooperate.