PYI: Theoretical and Experimental Studies of Animal FeedingBehavior

点击翻译按钮获取中文摘要

This Presidential Young Investigator award will provide support for theoretical and experimental analyses of feeding behavior, which Dr. Stephens views as a useful model for many behavioral systems. The research will focus on the economics of the feeding behavior of vertebrate animals in randomly varying environments. Most theoretical models in behavioral ecology estimate the fitness value of behavioral tactics in terms of some immediate gain (e.g., of food, mates, or offspring). The models ignore the gains of information that an option might provide; for example, a given food patch may provide both calories and information about the patch's relative quality. Dr. Stephens has devised a new way to estimate the value of information, and he will attempt to answer several new experimental and theoretical quesions suggested by this technique. Evidence from experiments on risk-sensitivity and on preferences for immediacy suggest that the temporal pattern of food acquisition is an important (but usually neglected) aspect of foraging. Dr. Stephens will attempt to show the theoretical relationships between risk-sensitivity and preferences for immediacy, and will attempt to distinguish experimentally among the many theoretical models that have been proposed to explain these two phenomena. Learning is usually viewed as an adaptation to an unpredictable environment. Dr. Stephens' theoretical work, in its earliest stages, suggests that learning is as much an adaptation to predictability as it is an adaptation to unpredictability. Further development of these theoretical models has a tremendous potential to change conventional views about animal learning and comparative learning abilities. Dr. Stephens' research might be applied in two practical contexts. An understanding of feeding behavior is fundamental to well-informed management of animal populations and ecosystems, and understanding how natural selection has solved behavioral decision-making problems may help students of human decision- making.