Diminished aversive classical conditioning in pathological gamblers

Aims  Impaired ability to form associations between negative events in gambling and aversive somatic reactions may be a predisposing factor for pathological gambling. The current study investigated whether a group of pathological gamblers and a control group differed in aversive classical conditioning. Design  A differential aversive classical conditioning paradigm, which consisted of three phases. In the habituation phase, one 850-Hz tone stimulus and one 1500-Hz tone stimulus were presented three times each in random order. In the acquisition phase, the two tones were presented 10 times each in random order, and one was always followed by a 100-dB burst of white noise. In the extinction phase the two tones were presented three times each without the white noise. Setting  University laboratory testing facilities and out-patient treatment facilities. Participants  Twenty pathological gamblers and 20 control participants. Measurements  Duration of seven cardiac interbeat-intervals (IBIs) following tone offset, gambling severity, tobacco and alcohol use, anxiety and depression. Findings  No group differences were found in the habituation and acquisition phases. However, a significant group × stimuli × trials × IBIs interaction effect was found in the extinction phase (P < 0.049). Follow-up analysis indicated that the pathological gamblers did not show aversive classical conditioning, but that the control group did. Conclusions  Pathological gamblers have a diminished capacity to form associations between aversive events and stimuli that predict aversive events. Aversion learning is likely to be an ineffective treatment for pathological gamblers.