Previous research suggests that mental health stigma creates significant barriers to treatment seeking and adherence, diminishes treatment outcomes, and motivates social rejection towards people experiencing mental illness; by contrast, compassion seems to offer protective effects, improving treatment outcomes and helping behavior. The current work extends the established literature by experimentally examining the independent and interactive effects of two factors theorized to influence stigma and compassion: controllability and language. Participants read vignettes about hypothetical mental illnesses explained with a genetic attribution (indicating low controllability) or a behavioral attribution (indicating high controllability) and completed measures of perceived controllability, stigma, and compassion. We found that genetic etiology, compared to behavioral etiology, decreased stigma and increased compassion. Although not statistically significant, preliminary evidence suggests that language might interact with
etiology to affect stigma. In the behavioral etiology condition, identity-first language (compared to person-first) exacerbated stigma, whereas, in the genetic etiology condition, this effect was descriptively reversed, though statistically nonsignificant. Our findings provide evidence that emphasizing the contribution of uncontrollable factors (e.g., genetics) to psychopathology could help reduce stigma and increase compassion for people experiencing mental illness. Language may also interact with controllability to inform stigma. This work could aid in advising empathetic and supportive language practices dependent on condition characteristics (e.g., perceived controllability), however, replication is needed to demonstrate the reliability of these effects.