Individual-based models are essential for sustainability science because they can represent how populations and communities are affected by complex and nonlinear environmental influences, feedbacks, and interactions. But for IBMs to be useful, they must adequately represent individual adaptive behavior: How do organisms make key decisions considering the current state of their environment and themselves? Classical behavior theory often is inadequate in realistic models where individuals must make tradeoffs among factors such as growth, risk, and reproductive output while environmental conditions are variable and affected by behavior of other individuals. In this session, speakers will present a variety of methods used to represent adaptive behavior in models of real ecological systems that contain enough complexity to be useful for management ecology. Speakers will describe the system they modeled and the adaptive decisions individuals make, their approach to modeling decision-making, and the extent to which the model succeeded in producing realistic dynamics at the individual and system levels. Approaches will include empirical models that reproduce observed behavior patterns, artificial evolution of neural nets and related decision rules, and “fitness-seeking” in which individuals use prediction and approximation to estimate which alternative offers highest future fitness. This session will be of value to ecologists interested in models that consider the effects of adaptive behavior on population and community ecology. It will be of special value to those interested in models of real sustainability and management issues in specific ecological systems, because many such problems can be addressed only with IBMs that consider behavior.