Meeting Abstract
Because genetic variation is required for adaptive evolution, it is essential to understand the mechanisms that maintain it in the face of selection. Although genetic errors such as mutation are primary, here we demonstrate another independent mechanism for maintaining genetic variation in populations: random epigenetic errors in development. We simulated populations of mobile, autonomous robots in which genomes encode morphological and neural structures, spatial relations, and regulatory elements; the interactions of structures and regulatory elements unfold in an explicitly modeled developmental process. Our system also explicitly models random genetic errors and random developmental errors. We simulated 11 levels of genetic error rate, 11 levels of developmental error rate, and their interactions in 10 populations of 60 simulated robots over 100 generations, with fitness determined by a simple movement task. In the presence of directional selection, genetic variation was proportional to the rate of random developmental error. Moreover, random developmental error and random genetic error are separate and independent mechanisms, as demonstrated by their statistical independence over evolutionary time. In addition, at all levels of developmental error, the mean individual fitness increased over generational time. This model therefore is consistent with the prediction that random developmental error is an evolutionary mechanism that maintains genetic variation, which, in turn, enables sustained adaptive evolution. This work was funded by the U.S. National Science Foundation (grant no. 1344227, INSPIRE, Special Projects).
