It has been shown recently that a specific intervention designed to modify the overgeneralization bias [i.e., concreteness training (CNT)] can decrease depressive symptoms. The intervention, however, involves multiple components (e.g., relaxation, problem solving), so it is not clear if the increase in concrete processing is the crucial mechanism in ameliorating depressive symptoms. Using a dismantling design, this online study examined whether targeting only concrete processing in the absence of a therapeutic context reduces depressive symptoms. Forty-two stable dysphoric participants were randomly allocated to either a waiting list or a 7-days concreteness training condition. Compared with the control group, concrete processing training resulted in a significant increase in the concreteness of thinking. No significant differences in autobiographical memory specificity, depressive symptoms, or rumination, however, were obtained post-intervention between the two groups. These findings suggest that concrete processing can be trained, but training effects may not generalize to untrained cognitive contents. The effectiveness of CNT as a standalone treatment for depression may be limited.