Trait anxiety and fear generalization: Overgeneralization of fear or undergeneralization of safety learning?

The tendency to overgeneralize fear learning has been identified as a potential risk factor for the maintenance of anxiety disorders. In this study, we used a morphed shape dimension to separately measure generalization of fear (learning when an aversive stimulus is present) and generalization of safety (learning when an aversive stimulus is absent) in participants classified as high or low in trait anxiety. In two experiments, we found that high trait anxious participants undergeneralized safety learning relative to low trait anxious participants, but there were no differences in overgeneralization of fear. A control group who learned about a neutral outcome did not show this pattern of results, and the aversiveness of the outcome predicted undergeneralization of safety learning, but did not predict overgeneralization of fear learning. Our findings suggest that overgeneralization of fear in clinically or high trait-anxious participants may be partly driven by undergeneralization of safety learning. Therefore, promoting generalization of safety learning may be a useful strategy to reduce the spread of fear.