Armstrong, B. C. . 1 , Dumay, N. 1, 2 , Kim, W. . 3 & Pitt, M. A. . 4
1 BCBL. Basque Center on Cognition, Brain and Language, Spain
2 Department of Psychology, University of Exeter, United Kingdom
3 Howard University
4 Department of Psychology,The Ohio State University
Connectionist accounts of quasiregular domains, such as spelling-sound correspondences in English, represent exception words (e.g., pint) amidst regular words (e.g., mint) via a graded ?warping? mechanism. Warping allows the model to extend the dominant pronunciation to nonwords (regularization) with minimal interference (spillover) from the exception. We tested for a behavioral marker of warping by investigating the degree to which participants generalized newly learned made-up words, which ranged from sharing the dominant pronunciation (regulars), a subordinate pronunciation (ambiguous), or a previously non-existent (exceptional) pronunciation. The new words were learned over two days, and generalization was assessed 48 hours later using nonword neighbors of the new words in a tempo naming task. The frequency of regularization (a measure of generalization) was directly related to degree of warping, thus substantiating the warping mechanism, and challenging alternative theoretical accounts of how exception and regular words coexist. The results also offer new insight into how domain-general learning mechanisms can, in some circumstances, exhibit stimulus-specific behavior, a paradox in statistical learning. More broadly, this work highlights the value of developing theories of representation that are integrally tied to how those representations are learned and generalized.