Proceedings of the Annual Meeting of the Cognitive Science Society

Labels facilitate infants category learning. Providing the same label for a set of distinct individuals enhances infantsability to identify the underlying category. In infants daily life, however, many category exemplars will go unlabeled, andinfants will inevitably receive a mix of labeled and unlabeled exemplars when learning real-world object categories. Here,we ask whether 2-year-old infants can integrate these labeled and unlabeled exemplars when learning a novel category.To do so, we draw on machine learning research in semi-supervised learning, a class of algorithms designed to learnfrom just such mixed data. Our results suggest infants do engage in semi-supervised category learning. Infants learnedcategories as successfully in a semi-supervised condition as in a fully-labeled conditionand more successfully than in anunlabeled condition. These findings reveal that the power of labels extends beyond the exemplars being labeled: labelingalso promotes infants learning from subsequent, unlabeled exemplars.