The effect of training infant joint visual attention on novel word learning.

Joint visual attention is the ability to follow another person\u27s line of vision. This ability develops toward the end of the human infant\u27s first year, but can be trained as early as 8 months of age. Many studies have correlated infant joint visual attention abilities with language development, but there has been no experimental manipulation to determine whether there is a causal relation between joint visual attention and word learning. In the present study, 41 infants between 7.5 and 8.5 months of age were assigned to either the treatment or control group. The treatment group received a joint attention training procedure, by which they were taught to follow the gaze of an experimenter to a toy through a three-phase conditioning process. The first phase was a baseline phase to determine whether the infant already had joint attention skills. The second phase was a learning phase during which the toy was activated by remote each time the experimenter looked at it. The final phase was a testing phase during which the toy was only activated if the infant looked to it after the experimenter did. The control group infants received a similar training, but were taught to look at a toy following presentation of a tone. Following training, all infants were taught two novel words within a joint attention context, in which the experimenter looked at a toy and labeled it with a novel word; this process continued four times for each word. Infants were then tested to determine whether they had learned the meaning of the new words. Of the 41 infants, only 5 in each group were trainable; 30 of the 41 infants learned at least one of the two novel words. Results indicate that there were no significant relations between joint visual attention training and novel word learning. This lack of findings may indicate that joint visual attention does not facilitate word learning. However, methodological limitations of this study (e.g., small sample size, few number of infants trainable, lack of variability in word learning) may have prevented potentially significant relations from being demonstrated