Alcohol consumption following bidirectional shifts in body weight in rats.

Weight restriction has frequently been used to induce consumption of pharmacologically significant amounts of alcohol by rats. When previously weight-restricted rats are fed ad lib., however, their alcohol consumption is substantially reduced. This could occur because weight restriction per se causes increased alcohol consumption, or because the stimulus conditions (in this case, largely interoceptive) that were originally associated with drinking have changed, resulting in a stimulus generalization decrement. The present experiment was designed to discriminate between these two possibilities. Two groups of animals, one at free-feeding weight (FFW) and one at 80% FFW were initially trained to drink alcohol in a limited access paradigm. Each group was then divided into two, such that one half remained at its original weight and the other was gradually shifted to the reverse feeding condition in a double cross-over design. When alcohol consumption was again stable the experimental groups were returned to their original weights. If the weight restriction hypothesis were true, animals should drink more when weight-reduced and less when at FFW. According to the stimulus generalization decrement hypothesis, drinking should decrease with any shift from their initial weight. The direction of the results was such as to support the weight restriction hypothesis, but the magnitude of the changes was greater than would be expected on the basis of the respective control groups, and the effects were often transient. In these ways the results more closely resembled positive and negative contrast effects that have traditionally been described following shifts in amount of reinforcement. Possible mechanisms and the importance of an animal's history as a determinant of alcohol consumption are discussed.