| The delayed response (DR) task has been used to investigate working memory (WM) in infancy. Much of the previous research on infant WM was designed to describe how WM abilities change with age. However, the identification of individual differences in infant WM is also of considerable interest.; The present study explored the emergence of WM and its relation to language late in the first year by following a group of 46 infants from 10 to 12 months. At each age, infants participated in two DR tasks: one that required a gaze response (Peek-a-Boo task) and one that required a reach response (Hide-and-Find task). Each task challenged WM durability across various delays (3, 9, or 15 seconds) and number of hiding locations (2 or 3). Infants also received two language assessments (Preschool Language Scale-4; The MacArthur CDIs). Test-retest reliability was assessed by testing infants with the same DR procedure at two separate visits.; Direction of initial gaze was the dependent variable in the Peek-a-Boo task, while in the Hide-and-Find task, the infant's first reach and direction of first gaze were coded. In the Peek-a-Boo task, 12-month-olds demonstrated superior performance compared to 10-montholds on 3-second trials. However, with 9-second trials, when the trials consisted of two locations, 10 and 12 month performance was nearly equivalent, while with three locations, 10 month performance was at chance and 12 month performance was below chance. At 12 months, infants performed better in the Hide-and-Find task compared to 10-month-olds. On trials where the direction of gaze and direction of reach were different, 10-month-olds were more likely to gaze correctly, while at 12 months this effect disappeared. There was some stability of individual differences in WM across age for the Hide-and Find task, but not for the Peek-a-boo task. Test-retest reliability calculations indicated low reliability of WM performance in both the Hide-and-Find and Peek-a-Boo tasks. Finally, there was a weak relation between individual differences in WM performance and individual differences in language, as measured by the PLS-4. Planned modifications to the procedures and directions for future research will be discussed. |
