Learning, Decision making and Evolutionary theory: Can we bridge the gap?
At the practical levels, learning processes are dynamically complex and are therefore difficult to model with analytical tools currently available for theoreticians in the field. This problem may partly be solved by the use of extensive computer simulations, but these have their own limitations. Thus, a practical challenge we now face is how to improve our ability to accommodate empirical knowledge on learning and decision making mechanisms when modeling the evolution of behavior.
The next challenge, which is even more difficult, is how to ensure that the selective forces acting on the behavioral trait under scrutiny are the same as those acting on the learning mechanisms that produce it. For example, the number of prey items that a starling should bring to the nest after each foraging trip may be considered as a behavioral trait that evolved to maximize food delivery rate. Yet, the learning and decision making mechanisms that serve this behavior might have evolved to serve a much broader set of behaviors, and modeling their adaptive value requires that we study all of them at once, which is usually impractical. Nevertheless, it would be unfortunate (and counter productive) if such difficulties were to discourage functional evolutionary approaches to the study of brain and behavior.

We believe that recognizing these problems can help to develop better tools and approaches to bridge the gap between "learning-absent" evolutionary studies of behavior, on the one hand, and "evolutionary-absent" mechanistic studies of learning and decision making, on the other. Some progress may be made by focusing on highly specific learning systems such as imprinting or song learning. Another direction might be to search for principles of learning likely to evolve convergently due to the structure of causality and to consider how they could be modified by different selection pressures and increase in complexity and diversity through evolution. The proposed workshop aims to raise and explore such ideas by a group of experienced empiricists and theoreticians from a range of disciplines (evolutionary biology, behavioral ecology, psychology, and computer sciences) who share an interest in learning and evolution.