ch they asked participants to generate as many uses as they could (i.e., do-your-best instructions) for a common object, regardless of the creativity of those uses. Pilot work with this task showed that participants felt that they knew how many uses they had generated (their output). Therefore, to manipulate the potential for self-evaluation, Szymanski and Harkins either provided an evaluative criterion by telling participants that at the end of the session they would be told the average number of uses generated by participants in a previous version of the experiment, or Szymanksi and Harkins withheld the criterion by telling participants that to preserve confidentiality they would not be given this information. When the participants were provided with the evaluative criterion, they had an opportunity for selfevaluation by social comparison. This manipulation of the potential for self-evaluation was crossed with a manipulation of the potential for experimenter evaluation. Half of the participants were told that at the end of the session, the experimenter would come in and count the number of uses that they had generated (experimenter evaluation), whereas the other half of the participants were told that the experimenter was interested in average performance so their individual outputs would be pooled with those of others and would not be examined (no-experimenter evaluation). Replicating the social loafing effect, Szymanski and Harkins (1987) found that participants subject only to experimenter evaluation (experimenter evaluation/no criterion) produced more uses than did participants who could be evaluated by no one. Consistent with the hypothesis that an opportunity for self-evaluation through social comparison would motivate participants, Szymanski and Harkins found that the potential for selfevaluation alone (no-experimenter evaluation/ criterion) also led to performance better than that found in the no-evaluation condition (i.e., no...
