Generalization of dimension-based statistical learning

Recent research demonstrates that the diagnosticity of an acoustic dimension for speech categorization is relative to its relationship to the evolving distribution of dimensional regularity across time and not simply to its fixed value along the dimension. Two studies examine the nature of this learning in online word recognition, testing generalization of learning across lexical contexts, and testing the extent to which variability in the training inventory affects learning. The results indicate that learning generalizes poorly across lexical contexts, but generalization may be boosted when listeners experience the dimensional regularity across multiple lexical items.