A new experimental paradigm to study children's processing of their parent's unscripted language input.

This paper introduces a new experimental paradigm for studying children's real-time language processing of their parents' unscripted speech. Focusing on children's processing of referential expressions, or the phrases that parents used to label particular objects, we engaged dyads in a game in which parents labeled one of several objects displayed on a screen, and the child was to quickly identify it as their eye gaze was tracked. There were two conditions; one included a competitor object (e.g., the target was a striped umbrella and the display also included an umbrella with polka dots), while the other one did not (e.g., only one umbrella was present). The results revealed evidence of children's incremental processing of their parents' referential expressions. They also showed faster processing of postnominally-modified as compared to prenominally-modified referential expressions. Parents tended to produce postnominally-modified referential expressions in the more difficult experimental condition in which there was a competitor object, suggesting either that these expressions are also easier for them to produce, or that they accommodate their children by producing more easily processed expressions. We discuss the potential of this paradigm for advancing theories of the relationship between child-directed language input and children's language processing.