Abstract: There have only been a few studies that tested the interaction between rumination and stress in predicting levels of depressive symptoms and results are mixed. Some of the possible explanations listed below. First, Treynor, Gonzalez, and Nolen-Hoeksema (2003) have identified two distinct subtypes of rumination - brooding and reflection - which explains the relationship between rumination and depression. More specifically, studies indicate that brooding, but not reflection, predicts the development of depressive symptoms over time. Second, most of the studies have employed cross-sectional designs, which preclude directly examining the interaction between rumination and stress. Accordingly, the aim of the current study was to examine whether rumination and its subtypes moderates the relationship between stress and depressive symptoms.
During the initial assessment, 618 high school students (302 boys, 316 girls) completed measures assessing stressful life events, rumination (including brooding and reflection) and depressive symptoms. Every three months for the next twelve months participants completed measures assessing stressful life events and depressive symptoms. Data were analyzed using hierarchical linear modeling analyses.
Hierarchical linear modeling analyses indicated adolescents who tend to ruminate in response to depressive symptoms did not report greater elevations in depressive symptoms following elevations in negative life events than other adolescents did. Additionally, brooding only showed the main effect on depressive symptoms and did not moderate the relationship between the occurrence of stressful life events and depressive symptoms, while reflection did not predict the development of depressive symptoms.
Stressful life events and brooding could predict the development of depressive symptoms, however rumination and its subtypes did not moderated the relationship between stress and depressive symptoms.

Key words: rumination, brooding, reflection, hierarchical linear model