Group cohesion, task performance, and the experimenter expectancy effect.

Studied the effects of cohesion on task fulfillment and explored the influence of task fulfillment on the initial level of cohesion. Within 4-person groups of undergraduates, cohesion was manipulated successfully by a triple procedure. The level of cohesion was ascertained directly after the induction treatment and again after task fulfillment. Group and individual ability tasks were used in Exp I (144 Ss) and Exp II (91 Ss), while individual learning tasks were used in Exp III (126 Ss). The hypothesis that cohesion facilitates task performance was partly supported for both individual and group tasks. All 8 comparisons of the means favored the high-cohesion condition; only 4 of the differences, however, were significant. Performing a group task tended to raise cohesion, whereas individual performance lowered liking for the group. There were no indications that the findings were a function of a pretest effect of the cohesion questionnaire or the experimenter's awareness of the research hypotheses. The facilitating influence of cohesion on task fulfillment is considered a consequence of heightened drive level of group members