Date of Award

Spring 1991

Project Type


Program or Major


Degree Name

Doctor of Philosophy

First Advisor

Victor A Benassi


According to the network model of mood and memory (Bower, 1981), information that is congruent with present mood should be primed and therefore more accessible than incongruent information. The following series of studies investigated this hypothesis and addressed several secondary predictions concerning the conditions under which mood-congruent priming effects occur. Using naming time as a measure of accessibility, Study 1 tested the mood-congruent priming hypothesis by assessing the relation between natural occurring mood and relative accessibility of positive versus negative adjectives and nouns. In addition to replicating Study 1, Study 2 tested an affect-specific version of the mood-congruent priming hypothesis that predicts that measures of specific mood (i.e., happy versus sad) should be related to the relative accessibility of happy versus sad associated words. Finally, based on a revised version of the network model of mood and memory, Study 3 investigated the hypothesis that the affective valence of mood-congruent words is more accessible than the affective valence of incongruent words. This hypothesis was tested by employing a valence judgment task in which subjects judged whether words were positive or negative. Although only Study 1 provides support for the general mood-congruent priming hypothesis, Studies 1 and 2 support the affect-specific mood-congruent priming hypothesis. Measures of happy-sad mood were associated with relative naming time for happy versus sad associated words such that the happier the mood the relatively shorter the naming time for happy as opposed to sad words. The implications of these results for automatic (i.e., noneffortful, nonmotivated) processes in mood-congruency findings are discussed. Contrary to the results of Studies 1 and 2, results of Study 3 support a mood-incongruency interpretation. Nouns incongruent with present mood were judged quicker than mood-congruent nouns. Although these findings dispute the revised version of the network model, potential problems with the judgment task are discussed along with suggestions for future research. Finally, several ancillary findings are reported, including mood- and depression-congruent recall and mood- and depression-congruent frequency of usage judgments.