Learning argument structure generalizations

Top Cited Papers
Publisher Website
Google Scholar
Abstract
General correlations between form and meaning at the level of argument structure patterns have often been assumed to be innate. Claims of in- nateness typically rest on the idea that the input is not rich enough for gen- eral learning strategies to yield the required representations. The present work demonstrates that the semantics associated with argument structure generalizations can indeed be learned, given the nature of the input and an understanding of general categorization strategies. Examination of an ex- tensive corpus study of children's and mothers' speech shows that tokens of one particular verb are found to account for the lion's share of instances of each argument frame considered. Experimental results are reported that demonstrate that high token frequency of a single prototypical exemplar facilitates the learning of constructional meaning.