Proceedings of the Annual Meeting of the Cognitive Science Society

The framework of cognitively bounded rationality treats prob-lem solving as fundamentally rational, but emphasises that itis constrained by cognitive architecture and the task environ-ment. This paper investigates a simple decision making heuris-tic, Take The Best (TTB), within that framework. We formu-late TTB as a likelihood-based probabilistic model, where thedecision strategy arises by probabilistic inference based on thetraining data and the model constraints. The strengths of theprobabilistic formulation, in addition to providing a boundedrational account of the learning of the heuristic, include naturalextensibility with additional cognitively plausible constraintsand prior information, and the possibility to embed the heuris-tic as a subpart of a larger probabilistic model. We extend themodel to learn cue discrimination thresholds for continuous-valued cues and experiment with using the model to accountfor biased preference feedback from a bounded rational agentin a simulated interactive machine learning task.