There was a psych study where each participant was supposed to press a button to make a light flash. The flashes were random. Depressed people were more likely

There was a psych study where each participant was supposed to press a button to make a light flash. The flashes were random. Depressed people were more likely to recognize this while non-depressed people were more likely to believe they had control. The vast extrapolation of this into "depressive realism" has always struck me as misguided. The test favors depressed people's instincts because they tend to have less belief in self-agency in general. If the light was influenced by the button you might have seen the opposite gap, with the non-depressed being more accurate.