Decisions and success of heterogeneous population of agents in learning to cross a highway

Abstract

We investigate the performance of a population of agents learning to cross a cellular automaton based highway. This performance is measured by mean values and their standard deviations of numbers of agents': (1) correct crossing decisions; (2) incorrect crossing decisions; (3) correct waiting decisions; (4) incorrect waiting decisions. Additionally, it is measured by mean values and their standard deviations of numbers of queued agents at simulation end. We study how agents' performance depends on the type of decision-making formula they use and on the presence of risk takers and of risk avoiders in the population of agents. We consider two decision-making formulas, one based on the assessment of both crossing and waiting decisions, and another one based only on the assessment of crossing decisions. We describe the simulation model focusing on the agents decision-making process and learning. The agents use an “observational social learning” strategy based on the observation of performance of other agents, mimicking what worked for them and avoiding what did not. Also, we investigate how accumulation of more information in agents' knowledge base affects agents' success in learning to cross the highway in homogeneous and heterogeneous populations of agents.