Choice and context: testing a simple short-term choice rule

Psychological studies of animal choice show that the immediate consequences of choice strongly influence preference. In contrast, evolutionary models emphasize the longer-term fitness consequences of choice. Building on recent work by Stephens & Anderson (2001, Behavioral Ecology 12, 330-339), this study presents two experiments that address this conflict. Stephens & Anderson developed an alternative choice situation based on patch-leaving decisions and compared this to the binary choice, or self-control, situation typically used in psychological studies. They hypothesized that the same short-term choice rule could account for choice in both situations, maximizing long-term gains in the patch situation, but typically producing shortsighted results in the self-control case. Experiment I used captive blue jays, Cyanocitta cristata, to test this 'same rule' hypothesis. The results do not support this hypothesis, suggesting that if a single rule applies, it is probably a more complex rule. Stephens & Anderson also hypothesized that a rule based on the delay to the next meal could explain why the intertrial interval has little effect in binary choice studies, even though the analogous travel time strongly affects patch-leaving decisions. When an animal leaves a patch, it experiences a delay consisting of the travel time plus time spent searching in the patch until food is obtained. Experiment 2 tested the hypothesis that travel time and search time combine additively, behaving like a single delay. Using treatments that created the same combined delay via different combinations of travel and search time, we found no evidence of nonadditivity, suggesting that these two components may indeed be treated as a single delay. (C) 2003 Published by Elsevier Ltd on behalf of The Association for the Study of Animal Behaviour.