Difficulties in acquiring verb meanings have motivated the syntactic bootstrapping theory of verb learning, which appeals to children's knowledge of syntax as well as their observations of events. But how might syntactic bootstrapping work? What relationships between sentence structure and meaning might be initially accessible to children? Recent findings have led to a revised version of syntactic bootstrapping, the analogical mapping view, in which little or no explicit syntactic knowledge is required. The central tenet of this revised view is that a representation of a sentence consisting of recognizable nouns ordered in a partial structure can be interpreted by analogy to a structured conceptual representation of an event. The proposed studies explore a strong prediction of the analogical mapping view: If children can interpret sentence structures by analogy, without prior syntax learning, then structure-sensitive interpretations of verbs should be found as soon as children can represent the structure of the sentence. To test this claim, proposed studies explore children's representations of sentences, asking under what circumstances the structural requirements of analogical mapping are met. A series of studies examines the role of repetition and priming in children's representations of speech. Does repetition in conversation allow very young children to represent multiword sentences accurately? Initial studies document a basic auditory word priming effect for 2- and 3-year-olds, and compare its properties to analogous findings with adults. Subsequent experiments follow up on this in three ways: First, the priming paradigm is used to explore preschoolers' representations of sentences, testing priming for word in sentences and for prosodic and sentence structures themselves. Second, findings on auditory priming will be compared to analyses of spontaneous speech to children, assessing how readily conversations display features which facilitate children's representation of sentences. Third, findings about children's representations of speech will be integrated into verb comprehension studies. Results from priming studies will guide the tailoring of discourse contexts for novel verbs to facilitate younger children's use of structural cues, and conversely to degrade older children's use of such cues. A more informed picture of young children's representations of input speech is crucial to testing hypotheses about the early interpretation of sentences. The priming paradigm provides a powerful tool for exploring young children's representations of language. Using this tool, the proposed studies examine how the repetition of linguistic structures supports the structure sensitive interpretation of sentences.
| Chambers, Kyle E; Onishi, Kristine H; Fisher, Cynthia (2003) Infants learn phonotactic regularities from brief auditory experience. Cognition 87:B69-77 |
| Onishi, Kristine H; Chambers, Kyle E; Fisher, Cynthia (2002) Learning phonotactic constraints from brief auditory experience. Cognition 83:B13-23 |
