Differences in rats and pigeons suboptimal choice may depend on where those stimuli are in their behavior system

Abstract

Timberlake (1993) proposed that much learning research can be better understood in the context of behavior systems theory. Learning theories generally have not considered how procedures may interact with evolutionarily prepared foraging contexts, thereby leading to anomalous conclusions. An example of such a conclusion is the apparent species difference when given a choice between a lower percentage of signaled reinforcement (20%) and higher percentage of unsignaled reinforcement (50%). Pigeons generally show a preference for the suboptimal alternative. Rats, however, often choose optimally. Orduña et al. have suggested that rats choose optimally because they show conditioned inhibition to signals for nonreinforcement, whereas pigeons do not. We suggest, however, that such fundamental species differences are unlikely. Rather, differences in the search mode elicited by the stimuli, together with some generalization between the signals for reinforcement and nonreinforcement are more likely to account for the species differences in suboptimal choice. When we used a retractable lever as the signal for reinforcement and a light stimulus as the signal for nonreinforcement, the rats showed suboptimal choice, not unlike that of pigeons. In this case, how the conditioned stimuli conform to the animals’ behavior systems natural predispositions appears to affect how the animals react to those stimuli. Thus, when the search modes are matched between species and there is no generalization between the two signals, similar behavior can be found. Timberlake’s contribution of behavior systems provides an evolutionary context in which species differences in sensory, response, and motivational differences can be separated from learning differences.