Proceedings of the Annual Meeting of the Cognitive Science Society

There is no guarantee that the set of possible theories that boundedly rational agents consider contains the true theory. And yet, these agents update their beliefs as new evidence comes in, leading to a conclusion about a particular domain. In this paper, we investigate under which conditions such agents arrive at sufficiently accurate beliefs compared to ideal agents. In doing so, we work within the framework of objective Bayesianism and draw on the literature on novel predictions in philosophy of science.