Proceedings of the Annual Meeting of the Cognitive Science Society

The Frame Problem (FP) is a puzzle in philosophy of mindand epistemology, articulated by the Stanford Encyclopedia ofPhilosophy as follows: “How do we account for our apparentability to make decisions on the basis only of what is relevantto an ongoing situation without having explicitly to considerall that is not relevant?” In this work, we focus on the causalvariant of the FP, the Causal Frame Problem (CFP). Assumingthat a reasoner’s mental causal model can be (implicitly) repre-sented by a causal Bayes net, we first introduce a notion calledPotential Level (PL). PL, in essence, encodes the relative po-sition of a node with respect to its neighbors in a causal Bayesnet. Drawing on the psychological literature on causal judg-ment, we substantiate the claim that PL may bear on how timeis encoded in the mind. Using PL, we propose an inferenceframework, called the PL-based Inference Framework (PLIF),which permits a boundedly-rational approach to the CFP, for-mally articulated at Marr’s algorithmic level of analysis. Weshow that our proposed framework, PLIF, is consistent withseveral findings in the causal judgment literature, and that PLand PLIF make a number of predictions, some of which arealready supported by existing findings.