The goal of Study 2 was to obtain additional evidence for the hypothesis that cognitive downward comparison may positively affect the perceived quality of the relationship. Moreover, we examined the extent to which the positive effect of downward comparison would occur in particular among individuals who were discontent with the partner and were strongly oriented to engage in social comparison. To examine the generalizability of the findings obtained in Study 1, Study 2 was conducted in a convenience sample of adults who had on average a relationship for more than 15 years.

To summarize, although there was no evidence that cognitive downward comparison exercised a main effect on relationship satisfaction and feelings of commitment to the relationship, a pattern of significant interactions demonstrated that the impact of downward comparison was moderated by discontent with the partner and by social comparison orientation. As we expected, cognitive downward comparisons improved the perceived quality of the relationship especially among people who were discontent with the relationship with their partner. Moreover, this buffering effect of cognitive downward comparison was moderated by social comparison orientation in such a way that only among those high in social comparison orientation was the negative impact of discontent on the perceived quality of the relationship reduced as a result of the downward comparison thought-generating task.

Also in accordance with the findings of Study 1, the number of the qualities generated did not differ between both conditions, thus suggesting that our findings were not artifacts of the number of qualities but that they were due to the effects of downward comparison. As a whole, the present findings suggest that even among individuals who had on average a relationship that had lasted over 15 years, cognitive downward comparison may help in maintaining a relative positive evaluation of one's relationship when experiencing relational discontent.

Nonetheless, it should be noted that one possible limitation of the results of Study 2 is that the main effect of cognitive set on perceived relationship quality was not replicated. The absence of an improvement in perceived quality of the relationship may be due to the fact that the pretest measures (i.e., social comparison orientation and discontent with the partner) and the dependent variables were assessed in the same session. The pretest measures may thus have influenced the subsequent dependent measures that were administered, known as "carryover" effects (see, for instance, Nunnally & Bernstein, 1994).

Additional limitations of Study 2 are that there was little control of the experimental situation, and that discontent was measured by only one item asking participants to indicate the degree to which they were discontent with the relationship with their partner. Therefore, a laboratory experiment was conducted to address these limitations and to further corroborate the findings of Studies 1 and 2.