Fifteen-month-olds can learn names of unseen objects in their absence

Human language permits us to call to mind representations of objects, events, and ideas that we cannot witness directly, enabling us to learn about the world far beyond our immediate surroundings. This essential capacity requires a referential link between words and representations of perceptually unavailable items. When and how does this link emerge? To address this question, we evaluated how early infants can learn names of unseen objects and identify those objects by name when they become visible. During training, 15-month-olds (N=67) and 12-month-olds (N=67) were primed to imagine a hidden object from a particular semantic neighborhood (e.g., fruits) and was labeled with a novel noun (e.g., “a modi”). During test, infants saw two novel objects – one from the primed neighborhood (e.g., a dragon fruit) and one unrelated object (e.g., an ottoman) – and heard the novel noun from training. We predicted a looking preference for the former object. Two control conditions evaluated the effects of semantic priming alone, regardless of the word (infants heard an entirely different novel noun at test) or any intrinsic looking preferences. Fifteen-month-olds looked to the object from the primed neighborhood upon hearing the word from training, but not an entirely different novel word, and showed no intrinsic looking preferences. Twelve-month-olds’ looking did not differ among conditions. This outcome held up even after seven days of vocabulary training designed to boost 12-month-olds’ lexicons. Our findings suggest that between 12 and 15 months, infants begin to represent unseen objects and learn their names.