How Do We Find Words In Implicit Artificial Language Learning?
Arnaud Destrebecqz, Cognitive Science Research Unit, Universite libre de Bruxelles, Belgium Axel Cleeremans, Cognitive Science Research Unit, Universite libre de Bruxelles, Belgium Implicit learning is often viewed as a central mechanism in natural language learning. In line with this idea, recent studies have shown that infants and adults could identify the “words” of an artificial language in which the only cues available for word segmentation are the transitional probabilities between syllables. However, the exact nature of the learning mechanisms and of the computational models that can account for these results remains controversial. According to one class of models, statistical learning amounts to parse the speech stream by forming chunks between adjacent elements. In this view, the sensitivity to statistical regularities is an emergent property following from the acquisition of rigid, conscious, word-like representations. According to a second class of models (the Sequential Rec
