Adaptive equations for fear and attraction

Background/Question/Methods The major body of ecological theory for animal behavior is based on the evolutionary premise of fitness maximization. Therefore, models of behavior are rooted in equations of fitness maximization, e.g. from life history theory, game theory, or state-dependent theory. Animals (as well as other life forms) are often in situations where decisions are influences by both internal state, age and the behavior of others. Therefore, optimization models based on only one of these aspects may not be sufficient for the richness of behaviors seen in nature.
An alternative approach has been to evolve behaviors and responses though a genetic algorithm module in an individual-based model (IBM) of a population. In most such cases the genetic algorithm (GA) has been used to evolve an artificial neural network (ANN) in the organism. However, it is hard to analyze the trade-offs and preferences in an ANN, and an equation set may therefore be advantageous.
This paper presents an individual-based model of a population where the behavioral decisions of each organism depends on the individual’s inherited fears and pleasures, which are given as “genes” in equations for responses to perceptions of food, light (proxy for risk and opportunity), temperature and conspecifics.

Results/Conclusions This method allows the organisms to respond simultaneously to age-, state-, and density-dependent ecological factors.
Irrespective of the starting point of the “genes”, the GA in the IBM drives the gene pool towards similar solutions. If neither growth nor mortality is density-dependent, the gene pool will evolve towards fixation of all “alleles”. However, in the density-dependent world, the gene pool will evolve to a spectrum of ways organisms will live, from risk-takers to risk-avoiders.
By the evolutionary force of the GA, the ecological realism possible to model in IBMs, and the focus on the proximate factors in the equations for fear and attraction, this modeling tool may act to bridge evolutionary ecology with more proximate and short-term perspectives in behavioral sciences.