Abstract

Negative affective bias, the tendency to prioritise the processing of negative relative to positive events, is commonly observed in depressed patients and is believed to be causally related to symptom development. However, why such biases develop is not known. Using an information theoretic framework, we investigated whether affective biases may reflect individuals’ estimates of the information content of negative relative to positive events.

In a series of studies, the information content of positive and negative task outcomes was manipulated independently by varying the volatility (unexpected uncertainty) or the noise (expected uncertainty) of their occurrence. Human participants altered the learning rates used for the outcomes selectively, preferentially learning from more volatile outcomes, but not adapting to the noise of the outcome. The volatility effect was associated with activity of the central norepinephrine system, estimated using pupilometry, for loss outcomes. In contrast, analysis using a lesioned Bayesian optimal model suggested that the apparent noise-blindness was due to an inability to track variations in the noise of outcomes across the task.

Humans maintain independent but imperfect estimates of the information content of distinct positive and negative outcomes which may bias their processing of affective events. Normalising affective biases using computationally inspired interventions may represent a novel approach to treatment for depression.