ABSTRACT
This work explores the interpretation and the computational modeling of discovery tasks—tasks where sudden insights result from cumulative information. Such tasks are useful in further understanding human everyday reasoning, beyond rule- based reasoning prevalent in cognitive science. We interpret the situation as involving mainly implicit memory with successive accumulation of information. We then implement this analysis in a cognitive architecture Clarion, which has succeeded in capturing a variety of human learning data in prior simulations. The simulation within this architecture accurately captures the human data. This work demonstrates the significant role played by implicit memory in human everyday reasoning. Furthermore, it demonstrates how such a reasoning process falls out of the existing cognitive architecture Clarion.
