The Role of "Explaining Away" in Human Abstract Rule Induction

Abstract

Of great interest to cognitive science is how human learning is constrained to avoid spurious generalizations. While many constraints must be relatively experience-independent, past experience provides a rich source of guidance for subsequent learning. If a learner discovers some structure in part of the environment, this can inform her future hypotheses about that domain. If a general structure parsimoniously accounts for particular sub-patterns, a rational learner should not stipulate separate explanations for each detail without additional evidence, as the general structure has “explained away” the original evidence. In a grammar-learning experiment using tone sequences, manipulating learners’ prior exposure to a tone environment affects their sensitivity to the grammar-defining feature, in this case consecutive repeated tones. Grammar-learning performance is worse if context melodies are “smooth”, that is, if small intervals occur more often than large ones, as this smoothness is a general property that accounts for a high rate of repetition.


Back to Table of Contents