Abstract: | ![]() The individual choice between conformity and innovation within task-oriented collectivities is presented as a social dilemma. Adaptive, network-embedded actors are seen to modify their propensities to conform or innovate retrospectively, based on performance differences between the individual and task group levels. A computational framework, based on simulation techniques and algorithmic complexity theory, is advanced to investigate the impact of antecedent structural conditions on innovative behavior and the effect that such behavior has on the evolution of patterns of interaction (conventions) and efficiency. Findings indicate that the dynamics of innovation differ dramatically based on the complexity of tasks faced by actors. When simple tasks are addressed, innovative behaviors have a destabilizing effect on social conventions and are clearly linked to contextual factors. When complex tasks are involved, commitment to innovation may actually help reinforce conventions and is not tied to structural antecedents.Department of Sociology, Stanford University |