When do children represent complex syntactic dependencies?

One core property of syntax is the capacity to encode abstract grammatical dependencies that can hold at a distance. When does this property emerge in development? We studied this question by testing infants’ representations of wh-questions, in which a fronted wh-phrase can act as the argument of a verb at a distance. For example, in What did the chef burn?, the wh-word what acts as the object of the verb burn, despite not appearing after the verb in canonical object position. Whereas prior work has focused on infants’ interpretations of these questions, we introduced a test to probe their underlying syntactic representations, independent of meaning. We asked when infants know that an object wh-phrase and a local object of a verb cannot co-occur because they both express the same argument relation. For example, What did the chef burn the pizza is not a well-formed question of English, because the pizza is acting as the object of the verb locally.

We tested 14- through 18-month-olds in a listening time preference task. We played infants grammatical and ungrammatical wh-questions in alternating trials, and measured how long they listened to each sentence type. We found that 18-month-olds differentiated between these sentence types, and treated them differently than sentences without wh-phrases: they preferred to listen to grammatical wh-questions that didn’t have local objects, but preferred to listen to grammatical declarative sentences with local objects (e.g., The chef burned the pizza). This tells us that they represent the wh-phrases as a non-local object of the verb in wh-questions. However, younger infants did not. These results suggest that the second year of life is a period of active syntactic development, during which the computational capacities for representing nonlocal syntactic dependencies become evident.