A potential pathway to the relapse of fear? Conditioned negative stimulus evaluation (but not physiological responding) resists instructed extinction

• Instructed extinction is an experimental analogue of cognitive interventions.
• Instructed extinction immediately reduced physiological indices of fear learning.
• Instructed extinction had no effect on ratings of conditional stimulus valence.
• This response dissociation is unlikely to reflect demand characteristics.
• The residual negative valence may provide a pathway for the return of fear.

Relapse of fear after successful intervention is a major problem in clinical practice. However, little is known about how it is mediated. The current study investigated the effects of instructed extinction and removal of the shock electrode on electrodermal responding (Experiment 1), fear potentiated startle (Experiment 2), and a continuous self-report measure of conditional stimulus valence (Experiments 1 and 2) in human differential fear conditioning. Instructed extinction and removal of the shock electrode resulted in the immediate reduction of differential fear potentiated startle and second interval electrodermal responding, but did not affect self-reported conditional stimulus valence. A separate sample of participants (Experiment 3) who were provided with a detailed description of the experimental scenario predicted the inverse outcome, reduced differential stimulus evaluations and continued differential physiological responding, rendering it unlikely that the current results reflect on demand characteristics. These results suggest that the negative valence acquired during fear conditioning is less sensitive to cognitive interventions than are the physiological indices of human fear learning and that valence reduction requires extended exposure training. Persisting negative valence after cognitive intervention may contribute to fear relapse after successful treatment.