Classical conditioning in humans: nicotine as CS and alcohol as US

Abstract

Animal studies suggest that drug effects can act as conditioned stimuli for various unconditioned stimuli including the effects of other drugs. The current study investigated drug-drug conditioning in human subjects. Sixteen subjects were given subcutaneous injections of either nicotine or saline before consumption of an alcoholic or soft drink in each of eight sessions. Across sessions the content of the injections was established as a reliable predictor of the alcoholic content of the drink. Physiological, subjective, and behavioural responses to the injections were used as indices of conditioning. Skin conductance measures obtained following the injections changed across trials in a way consistent with a conditioned response though patterns of change on cardiac inter-beat interval were less clear. However, neither behavioural nor subjective measures showed conditioning effects. In view of the number of variables studied the evidence for the development of conditioned responding on physiological measures must be suspected of being a type I error and is in need of replication. Subjects' reports revealed that nicotine and saline injections were difficult to discriminate. This would have weakened conditioning effects. Suggestions are made for improvements in the design of future studies of drug-drug conditioning in human subjects.