Establishing Discriminative Repertoires in the Evolutionary Theory of Behavior Dynamics: Preliminary Outcomes and Future Directions

The Evolutionary Theory of Behavior Dynamics (ETBD) has garnered significant empirical support in producing artificial organisms that engage in behavior patterns that have striking similarities to those of living organisms. However, the algorithm is not currently designed to generate stimulus control. Thus, the purpose of the present paper was to use a theory-driven approach in proposing some modifications to the ETBD to extend its application to discriminative repertoires in pursuit of stimulus control. The algorithm was first modified to initialize separate populations of behavior corresponding to the SD and SΔ conditions. Then, we allowed for learning carryover between the two schedules. Finally, the algorithm mutated previously emitted behavior out of the SΔ population to simulate extinction. The patterns produced by the modified algorithm appear conceptually consistent with how organisms learn, but future research needs to compare our results with those of living organisms to validate our propositions.