Imagery-based fear conditioning enhances the late positive potential

Abstract Pairing a conditioned stimulus (CS) repeatedly with aversive imagery as unconditioned stimulus induces conditioned subjective and autonomic fear responses, which can explain fears without prior aversive experiences. To investigate underlying brain dynamics, we assessed the late positive potential, an electrocortical response sensitive to classically conditioned fear, together with skin conductance, cardiac interbeat interval, and fear ratings in 48 participants. Participants underwent either 1) imagery-based fear conditioning, in which aversive CS+ (CS+ av ), neutral CS+ (CS+ neu ), and CS− were paired with cues to imagine an aversive electric shock, a mild vibration, or nothing, respectively, or 2) classical fear conditioning in which CS were paired with physical shocks, physical vibrations, or nothing. Similar to the classical conditioning group, the imagery-based conditioning group reported higher fear for CS+ av than for both nonaversive CS and displayed increased late positive potential amplitudes for CS+ av compared with CS− and with CS+ neu . Moreover, the magnitude of these effects did not differ significantly between imagery-based and classical conditioning. Meanwhile, strong cardiac deceleration and increased skin conductance responses to CS+ av were uniquely observed for classical conditioning. These results suggest that imagery-based conditioning evokes similar cortical and subjective, but not autonomic, responses as conditioning with physical unconditioned stimuli—advancing our understanding of fear in the absence of aversive real-life experiences.