Considering the Source: Preschoolers (and Adults) Use Talker Acoustics Predictively and Flexibly in On-Line Sentence Processing

The identity of the person talking is likely to constrain the things that they talk about. Adults can use talker acoustics to make on-line predictions about upcoming spoken material (Van Berkum et al., 2008). However, this cue to meaning may take time to learn. Do preschoolers consider who is talking when they are comprehending spoken sentences? I explored this question in two eye-tracked picture selection experiments. Experiment 1 showed that children and adults use vocal cues to talker identity in predicting the color of upcoming referents in spoken sentences. Experiment 2 showed that children and adults flexibly use acoustic cues to talker for first-person requests (“I want the square”) but reference to individuals for third-person requests (“Billy wants the square”). This suggests that children aged 3-5 years use who is talking to constrain the scope of reference in sentence processing, and know when this cue is likely to be useful.