Proceedings of the Annual Meeting of the Cognitive Science Society

When two constructions apparently serve the same function, which one is chosen by the speaker? We hypothesize that context determines the speaker's choice, and conducted an experiment where participants were asked to choose between the passive or the inchoative constructions after reading a short context. We operationalized their functional differences in terms of agentivity and predicted that speakers would choose the passive in high agentivity contexts, but would favor the inchoative in low agentivity contexts. There were 16 critical items, each presented in two conditions (high x low agentivity). 28 filler items were included, 8 of which were used as catch trials. A logistic mixed-effects model was fitted and the results confirmed our predictions, allowing us to conclude that speakers are aware of the constructions' functions, choosing the ones that fit best in context.