Explicit instructions regarding stimulus-threat associations increase acquisition and extinction of fear in healthy participants. The current study aimed to investigate the effect of contingency instructions on fear acquisition and extinction in patients with anxiety disorders. Patients with various anxiety disorders (N = 104) and healthy comparison participants (N = 93) participated in a differential fear conditioning task (within-subjects design). Approximately halfway through the acquisition phase, participants were instructed about the stimulus-threat association, and approximately halfway through the extinction phase, participants were informed that the unconditioned stimulus (US) would no longer be administered. Outcome measures were: fear-potentiated startle, skin conductance, fearfulness ratings, and US expectancy ratings. Patients demonstrated overall increased physiological and subjective fear responses during acquisition and extinction phases, relative to the comparison group. There were no major differences in fear acquisition and extinction between patients with different anxiety disorders. During acquisition, instructions led to increased discrimination of fear responses between a danger cue (conditioned stimulus [CS]+) and safety cue (CS−) in both patients and comparison participants. Moreover, instructions strengthened extinction of fear responses in the patient and comparison group. Patients and healthy comparison participants are better able to discriminate between danger and safety cues when they have been explicitly informed about cues that announce a threat situation. Considering the analogies between fear extinction procedures and exposure therapy, this suggests that specific instructions on stimulus-threat associations during exposure therapy might improve short-term treatment efficacy. The question remains for future studies whether instructions have a positive effect on extinction learning in the longer term.