Bridging Levels from Individuals to Communities and Ecosystems: Including Adaptive Behavior and Feedbacks in Ecological Theory and Models

Abstract

Symposium motivation and objectives Ecologists now recognize that the classical theory of population and community ecology is severely limited by its inability to represent the effects of individual adaptive behavior, but we are still looking for ways to accommodate adaptive behavior in models and theory. Conflicts between classical theory and adaptive behavior were brought to the forefront by analyses such as that of Abrams (1993), which showed that the fundamental assumptions of classic predator-prey trophic models are invalid when predator-avoidance behaviors are considered. Such analyses motivated the many empirical studies that documented and quantified such behaviors and their effects, for example, "trait-mediated indirect interactions" and "nonconsumptive effects," which have been a major focus of ecology since the 1990s. At the same time, individual-based models (IBMs; also referred to as agent-based models) were recognized as a way to incorporate adaptive individual behavior and its effects in ecological models and theory. The landmark paper of Huston et al. (1988) anticipated the contributions of IBMs to new theory that unifies behavior at the individual level with population and community ecology. However, even though IBMs are widely used in ecology and many other sciences, the anticipated new theory has been slow to develop. Grimm (1999) noted that IBMs rarely addressed theory either by developing new theory or by re-evaluating existing theory. But the 2000s did see the establishment of one new approach for developing new theory that explicitly unifies individual and higher levels of ecology: the "pattern-oriented theory development cycle" (Grimm and Railsback 2005, Grimm et al. 2005) in which theory for how individuals make adaptive decisions is tested by how well it reproduces observed patterns at both individual and higher levels. This approach poses alternative hypotheses for individual behaviors and falsifies those that, in an IBM, do not reproduce important and realistic phenomena of population and community ecology. The hypothesis-testing cycle can produce (and has produced) general, reus-able theory that explains higher-level ecology from individual behavior. Such theory is referred to as