Date of Award

Winter 1981

Project Type


Program or Major


Degree Name

Doctor of Philosophy


Second-order schedules, in which brief stimuli replace some food presentations, have been used to differentiate between physiological and psychological factors which might contribute to the induction of excessive drinking during exposure to periodic food schedules. Post-stimulus drinking is regarded as reflecting the influence of psychological factors.

In preliminary research, post-stimulus drinking was typically less reliable than post-food drinking. Rats which drank most reliably during post-stimulus periods also often resumed drinking after lever-pressing during post-food periods. Such drinking, like post-stimulus drinking, consisted of short bouts which alternated with bouts of lever-pressing. It was hypothesized that short-bout licking might be a terminal activity, one whose rate of occurrence varies directly with the probability of food presentations.

In Experiment 1, a contingency which required that licking occur during each component of a second-order schedule increased the rate of occurrence, but not the durations, of short lick bouts.

In Experiment 2, auditory signals informed animals, throughout each component, of the outcome (food or stimulus presentation) to follow the component. A clear discrimination in lever-pressing between pre-food and pre-stimulus signals was observed. Signal effects on drinking were, however, minimal. Levels of short-bout drinking exhibited by 8 of 9 animals presented with signals greatly exceeded those among 3 control animals not presented with signals. Such drinking declined following elimination of the signal-outcome correlation.

In Experiment 3, drinking after a food-paired stimulus on probe trials differed from that observed following presentations of a stimulus not paired with food. Drinking following the food-paired stimulus more closely resembled post-food drinking in bout duration, latency, and number and distribution of licks within components.

Results suggest that short-bout drinking seen during training with second-order schedules is independent of a response requirement. Such drinking can be brought under the control of reinforcement contingencies, but is not, in general, a terminal activity. It may, however, be sensitive to environmental predictability. The behavior pattern commonly described as schedule-induced drinking depends upon presentations either of food or of a food-correlated stimulus.