Using self-harm Implicit Association Tests (IATs), we sought to test whether (1) suicidal adolescents show implicit identification with self-harm and whether (2) IATs are reliable and sensitive to psychiatric change and (3) predict future suicide attempts. We administered 6 self-harm IATs to 71 adolescents from a psychiatric inpatient unit and assessed suicidal behaviors at admission, discharge and 3 months after discharge. Results were in the expected direction for each IAT but not statistically significant. After aggregating trials across IATs, suicide attempters showed increased implicit identification with self-harm, compared with non-suicidal controls. IATs showed good reliability and sensitivity to psychiatric change but did not prospectively predict suicide attempts. Adolescent suicide attempters may have stronger implicit associations with self-harm than non-suicidal controls.
Keywords: IAT; implicit association; prediction; suicide.