Measuring self-referent memory bias as marker for depression: overview, new insights, and recommendations

Negative self-referent memory bias (the preferential memory for negative self-referent information) is a well-known symptom of depression and a risk factor for its development, maintenance, and recurrence. Evidence shows its potential role as an add-on tool in clinical practice. However, an overview of the possible different outcomes is lacking and it remains unclear which self-referent memory bias measure(s) could be clinically relevant. Here, as a first step towards clinical implication, we investigate which measures best differentiate current depression status and track depressive symptom severity most closely. The total sample (N = 956) from three clinical datasets was divided into a current depression, remitted depression, and non-disordered control group. Self-referent memory bias task measures were calculated and the drift diffusion model (DDM) was applied to assess underlying components of the cognitive self-referent decision making process. Measures were compared between groups and linear regression models applied to assess their association with depressive symptom severity. The number of negative endorsed words differentiated best between depression status while a combination of the number of positive endorsed words, self-referent negative memory bias, and positive drift rate was most strongly associated with depressive symptom severity. Our results give direction to the clinical implementation of this task. Its value in assessing, monitoring, and predicting depressive state and trait in clinical settings requires further investigation.