<p>Previous studies have raised the question whether clinical conditions like depression may affect symbolic behavior. This study investigated whether participants with high and low levels of self-reported depressive symptoms differed on equivalence class formation, transformation of function, and maintenance. Participants were divided to two groups based on scores on the Beck Depression Inventory–II: high level and low level of self-reported depressive symptoms. Participants underwent matching-to-sample procedures to establish three equivalence classes, that included faces expressing emotions and abstract stimuli. The semantic differential and the implicit relational assessment procedure (IRAP) were then administered to assess the transformation of function. Two semantic differential control groups, also categorized by high and low level of self-reported depressive symptoms, evaluated the experimental stimuli using the semantic differential. A follow-up session occurred 30&#xa0;days later to verify the maintenance of results. The high-level group was less accurate in matching-to-sample training and tests compared to the low-level group. Transformation of function was similar for both groups, as measured by the semantic differential. However, a “happiness superiority effect” was observed only for the low-level group in the IRAP. Maintenance was similar for both groups, except for the evaluation of the happy stimuli using the semantic differential, which deteriorated for the high-level group. This was the first study to compare equivalence-like behaviors in depressed and nondepressed participants, using stimuli with emotional valence.</p>