The Effects of Animacy and Syntax on Priming: A Developmental Study.

Abstract

Sentence production relies on the activation of semantic information (e.g., noun animacy) and syntactic frames that specify an order for grammatical functions (e.g., subject before object). However, it is unclear whether these semantic and syntactic processes interact and if this might change over development. We thus examined the extent to which animacy-semantic role mappings in dative prime sentences and target scenes influences choice of syntactic structure (structural priming, analysis 1) and ordering of nouns as a function of animacy (animacy noun priming, analysis 2) in children and adults. One hundred forty-three participants (47 three year olds, 48 five year olds and 48 adults) alternated with the experimenter in describing animations. Animacy mappings for themes and goals were either prototypical or non-prototypical and either matched or mismatched across the experimenter's prime scenes and participants' target elicitation scenes. Prime sentences were either double-object datives (DOD e.g., the girl brought the monkey a ball) or prepositional datives (PD e.g., the girl brought the ball to the monkey), and occurred with either animate-inanimate or inanimate-animate, post-verbal noun order. Participants' target sentences were coded for syntactic form, and animacy noun order. All age groups showed a structural priming effect. A significant interaction between prime structure, prime animacy-semantic role mappings and prime-target match indicated that animacy could moderate structural priming in 3 year olds. However, animacy had no effect on structural priming in any other instance. Nevertheless, production of DOD structures was influenced by whether animacy-semantic role mappings in primes and target scenes matched or mismatched. We provide new evidence of animacy noun order priming effects in 3 and 5 year olds where there was prime-target match in animacy-semantic role mappings. Neither prime animacy noun ordering nor animacy-semantic role mappings influenced adults' target sentences. Our results demonstrate that animacy cues can affect speakers' word order independently of syntactic structure and also through interactions with syntax, although these processes are subject to developmental changes. We therefore, suggest that theories of structural priming, sentence production, linguistic representation and language acquisition all need to explicitly account for developmental changes in the role of semantic and syntactic information in sentence processing.