Input effects within a constructionist framework

Abstract

Constructionist approaches to language hypothesize that grammar can be learned from the input using domain-general mechanisms. This emphasis has engendered a great deal of research—exemplified in the present issue—that seeks to illuminate the ways in which input- related factors can both drive and constrain constructional acquisition. In this commentary piece, we situate results reported by contributors to the present issue within the larger body of acquisition work in the constructionist framework. We address the importance of both type frequency and skewed input samples in the development of constructional categories and we compare different ways that the association between verbs and constructions can be measured, including through the use of conditional probabilities, lexical biases, and introspective judgments.