A Rational Account of Categorization Based on Information Theory
This work addresses the problem of understanding human categorization for cognitive science, but it appears incremental as it builds on prior models without claiming broad SOTA or foundational impact.
The authors tackled the problem of explaining human categorization behavior by proposing a new information-theoretic rational analysis theory, and found that it performs at least as well or better than several existing models in accounting for key experimental findings.
We present a new theory of categorization based on an information-theoretic rational analysis. To evaluate this theory, we investigate how well it can account for key findings from classic categorization experiments conducted by Hayes-Roth and Hayes-Roth (1977), Medin and Schaffer (1978), and Smith and Minda (1998). We find that it explains the human categorization behavior at least as well (or better) than the independent cue and context models (Medin & Schaffer, 1978), the rational model of categorization (Anderson, 1991), and a hierarchical Dirichlet process model (Griffiths et al., 2007).