There is substantial evidence supporting the hypothesis that cognitive reactivity is an important variable in the etiology of depression. However, there is a lack of studies examining possible mechanisms that underlie cognitive reactivity. The present study tested whether two specific mood regulation processes differentially appear in vulnerable and non-vulnerable individuals, and whether they can account for differences in cognitive reactivity. In a cross-sectional experimental design, 20 formerly-depressed individuals (FD) were compared with 20 never-depressed individuals (ND). In an autobiographical memory task both groups differed concerning the use of positively and negatively toned emotion words: FD retrieved fewer positive emotion words than ND in the second phase of this task. Furthermore, FD with a high cognitive reactivity retrieved more negatively toned emotion words. In the ND group there was a different pattern: Subjects with a high cognitive reactivity retrieved less positively toned emotion words. Two different cognitive processes seem to account for cognitive reactivity in individuals who are at high versus low risk for depression.