The pattern of findings obtained in Study 5 thus supported the results of Study 4. That is, Study 5 replicated and supplemented the measures as used in Study 4 with measures that were expanded with additional items along with the addition of a measure of identification with the discussant. As in Study 4, an interaction of direction of comparison and target under evaluation was found: Consistent with the main findings of Study 4, it was seen as less socially inappropriate and undesirable to describe one's own intimate partner in an enhancing fashion than to describe oneself (as a partner) as positive and superior to others, whereas a partner-deprecating evaluation was rated as more socially inappropriate and undesirable than a self-denigrating evaluation.

In addition, Study 5 revealed that the differences in social desirability ratings were mediated by the degree of sympathy that people felt for the discussant. This finding suggests that perceptions of social desirability are driven by participants' feelings of sympathy for the discussant. Consequently, in response to the discussant's descriptions of the self or the partner individuals experience greater or lesser feelings of sympathy for the discussant which, in turn, will affect perceptions of social desirability of the self- and partner-evaluations. This was not the case for identification; the effects on identification appeared to be mediated by perceptions of social desirability, thus indicating that the effects on feelings of identification were mediated by individuals' perceptions of social desirability of the verbal statements.

Furthermore, participants in Study 5 were induced to evaluate the self or the partner in either an enhancing or in a deprecating way. As predicted, participants who were asked to evaluate their intimate partner in an enhancing manner were able to generate more qualities than those who were asked to evaluate the self (as a partner) in a like manner. This finding makes sense, if, as suggested earlier, partner-enhancement is considered as more socially acceptable than self-enhancement and people act accordingly.

Assuming that this is the case, people may anticipate possible negative reactions and may feel a greater reluctance to evaluate themselves in an enhancing fashion than to engage in partner-enhancing evaluations. An alternative explanation for the fact that people were better able to engage in partner- rather than in self-enhancing evaluations may be a differential cognitive availability of information about the self or the partner. Individuals who are involved in an ongoing close relationship probably have available to them more extensive positive mental models of their partner than of themselves (as a partner). Thus, it seems likely that this differential availability can affect the evaluation process of the partner or the self.

The first two studies were therefore successful in showing individuals' reactions to others' self- and partner-evaluations in either enhancing or deprecating terms. However, these studies did not investigate whether it was explicit social comparison that accounted for the effects on social desirability and recipients' feelings or the fact that the discussant made boastful statements regardless of whether or not he or she engaged in social comparison.

That is, the former studies did not examine the differential effects of comparative (i.e. with explicit references to social comparison) and noncomparative evaluations (i.e. without references to social comparison), a limitation that will be remedied in the next study. Study 6 was designed to confirm and extend the results of Studies 4 and 5 under the same carefully designed experimental conditions, manipulating systematically the comparative nature of self- and partner-enhancing evaluations and looking at a wider variety of possible responses in order to obtain further evidence for the predicted reactions to such downward social comparisons.