| When learning words, young children attend to cues such as a speaker's direction of gaze or goal directed behavior. The current experiments demonstrate that preschoolers also consider cues to a speaker's visual perspective, knowledge, and ignorance when determining the target of a referential expression. Experiment 1 tested four-year-old children's ability to identify the referent of a novel word when the speaker's eye gaze was directed at an incorrect object. The speaker fixated on a visible novel object while asking, for example, "Where's the nurmy?" Children were able to use the speaker's visual access and knowledge about the location of objects to correctly identify the hidden referent of a novel word, instead of merely following the speaker's eye gaze. In Experiments 2 and 3, children were again asked, "Where's the nurmy?" but in this case, both novel objects were hidden to the speaker. On test trials, the speaker knew the location of only one of the two objects because she had placed it in its final, hidden location. Both four-year-olds (Experiment 2) and two-year-olds (Experiment 3) succeeded at this task, selecting the object that was hidden without the speaker's knowledge. In a fourth experiment, a speaker used a new adjective while either touching or looking at an object with a novel appearance and a novel texture. Although it did not reach significance, 4-year-old children showed a trend towards using the speaker's perceptual access to determine which property (its appearance or its texture) was described by the new word. Together, these results suggest that children as young as two years of age track what a speaker does and does not know, and that they use this information when determining the referent of a novel word. |
