In two experiments, we investigated the relationship shared by irrelevant thoughts, emotional mood states, and cognitive task performance. At an empirical level, irrelevant thoughts were defined as thoughts that did not facilitate successful task performance. We used the same general procedure for both experiments: three groups of college students received happy-, neutral-(control), or sad-mood inductions and performed a memory task. The procedure for obtaining thoughts varied between experiments. The subjects in Experiment 1 listed their thoughts after the memory recall task. In Experiment 2, the subjects were tape-recorded while performing a memory task and producing concurrent verbal protocols. The subjects in both experiments then judged their thoughts in terms of frequency, intensity, and irrelevance. We found a similar pattern of results in both experiments: (l) The proportions of irrelevant thoughts and recall performance were negatively related, and (2) happy and sad students produced reliably greater proportions of irrelevant thoughts than did neutral (control) students.Every day, people experience emotional mood changes that influence performance. Yet the underlying dynamics of these effects are not clear. Emotional states have been shown to influence cognitive processes in a variety of ways (Alloy & Abramson, 1979;Anooshian & Siegel, 1985; Bower & Mayer, 1989; Clark, Milberg, & Erber, 1988;Ellis & Ashbrook, 1989;Ellis, Seibert, & Herbert, 1990;Forgas & Bower, 1988;Isen, 1984;Kenealy, 1986; Ucros, 1989). These investigations have shown that emotional mood states are related to (1) alterations in social and personal judgments, (2) alterations in spatial judgments, (3) mood congruence effects, and (4) recall impairment. The latter effect, recall impairment, will be the focus of the present investigation. Impairments in recall performance are observed when a complex task occurs in the context of relatively intense emotional mood states (Ellis, 1985;Hasher & Zacks, 1979).One way to study the effects of emotional states is to study thoughts, because thoughts vary in harmony with emotional mood states. Thoughts associated with emotional states can inspire or impair performance. These thoughts are often self-referent in nature. leagues (Sarason, 1984; Sarason, Sarason, Keefe, Hayes, & Shearin, 1986;Sarason & Stoops, 1978) found that subjects who were anxious about tests experienced cognitive interference that was peculiar to emotional states associated with test anxiety. These anxious states included thoughts pertaining to increased social anxiety or worry , increased intensity of fear of failure, and increased negative self-evaluation. These studies showed that while taking exams, individuals who were anxious about tests reported greater numbers of thoughts related to worrying about their performance than did nonanxious individuals. Thus, it appears that anxious people engage in thoughts that are qualitatively and quantitatively different from those of nonanxious people, and that these thoughts are relate...