Abstract
Visual short-term memory (VSTM), a subcomponent of working memory, is a memory system that briefly stores and maintains visual information across disruptions that may occur with eye movements and eye blinks (Luck, 2007). Research using the simultaneous stream change detection task has examined infants’ VSTM with arrays of as many as 6 items (Ross-Sheehy et al., 2003). More precise one-shot change-localization tasks have primarily tested infants’ VSTM for arrays of two items (Oakes et al., 2013). Here, we assessed VSTM in a group of 54 5- to 12-month-old infants using an eye-tracking change localization procedure. Here, infants saw trials with the following sequence: a 500-ms sample array of three or four (set sizes) colored circles, followed by a 300-ms delay array with a blank screen, and finally a 2000-ms test array in which one circle chosen at random changes color. We calculated change localization scores during the test phase for each infant on each trial by dividing the duration of infants’ looking at the changed item by their looking at all items (3 or 4 items) combined. Successful detection and localization of the changed item is indicated by above chance change localization scores, .33 for set size 3, and .25 for set size 4. Infants’ mean change localization was significantly above chance for both set sizes three, t(53) = 3.45, p = .001, and four, t(53) = 5.28, p < .001. In addition, change localization at set size three was significantly correlated with age, r = .30, p = .03, but not set size four, r = .12, p = .40, indicating increased sensitivity to change at set size three during infancy. These results add to our growing understanding of the development of VSTM in infancy, and demonstrate the utility of our change localization task with larger set sizes.