Abstract

The current study examined the cognitive heuristics employed by rats evaluating rewarding brain stimulation. Rats implanted with electrodes aimed at the lateral hypothalamus were trained to self-administer brain stimulation. In Experiment 1, one type of stimulation train was presented at a time: some trains were set at a constant frequency, whereas others were composite-frequency trains in which the frequency at the end of the train was set to be lower than in the preceding component. If the "peak" and "end" reward values were used as exemplars in evaluating the stimulation, the weaker end should degrade the overall reward value of the composite trains. However, the performance curves for all trains were overlapping, suggesting that all trains were equally rewarding. A model in which the "peak" exemplar alone is employed, is sufficient to account for these findings. In Experiment 2, constant- and composite-frequency trains were presented simultaneously. Adding a weaker end had no effect on preference, again suggesting that the simple "peak" alone model was sufficient. Experiment 3 was similar to the first experiment, in that constant- and composite-frequency trains were presented one at a time. However, the duration of the weaker, terminal portion of the composite trains was now greater. The performance curves for all trains were again generally overlapping, indicating no consistent preference for any train type. This result is again consistent with the "peak-only" model. The aim of Experiment 4 was to assess whether the "beginning" exemplar played a role in the evaluation of rewarding stimulation. Preliminary results did not support this hypothesis. Lastly, Experiment 5 examined whether duration could serve as an exemplar when the evaluation context was changed. In an earlier experiment involving trains presented one at a time, rats were indifferent to train duration increases once the stimulation exceeded 2-4 s. In Experiment 5, when comparable train durations were presented simultaneously, the longer train was more effective than the shorter trains in 4 out of 7 subjects. The implications of the study findings and future directions are discussed.