The Development of Visual Short-Term Memory Capacity in Infants Shannon Ross-Sheehy, Lisa M. Oakes, and Steven J. Luck Four experiments assessed visual short-term memory capacity in 4- to 13-month-old infants by comparing their looking to changing and nonchanging stimulus streams presented side by side. In each stream, 1 to 6 colored squares repeatedly appeared and disappeared. In changing streams, the color of a different randomly chosen square changed each time the display reappeared; the colors remained the same in nonchanging streams. Infants should look longer at changing streams, but only if they can remember the colors of the squares. The youngest infants preferred changing streams only when the displays contained one object, whereas older infants preferred changing streams when the displays contained up to 4 objects. Thus, visual short-term memory capacity increases significantly across the first year of life. The short-term storage of information is important for acquiring new knowledge, solving problems, and acting on current goals. For example, effective comparison and categorization of items that cannot be simultaneously foveated requires visual short- term memory, and visual short-term memory is therefore critical for the online use of visual infor- mation. The present article examines the develop- ment of visual short-term memory capacity in infants, with an emphasis on understanding how infant short-term memory abilities are related to adult short-term memory abilities. We begin by discussing the general role of short-term memory in cognition and then turn to the development of short- term memory. Short-Term Memory and Working Memory Psychologists have distinguished between short- term and long-term memory systems for more than a hundred years (e.g., James, 1890). The common view of short-term memory systems is that (a) they can create memory representations rapidly, (b) they can store only a handful of representations at any one time, and (c) they require active rehearsal to maintain the representations beyond a few seconds. In contrast, the common view of long-term memory systems is that (a) the memory representations are formed slowly, (b) storage capacity is essentially unlimited, and (c) the memories suffer from little or no decay of information. Research in the 1960s and early 1970s failed to provide compelling evidence for separate short- and long-term memory systems (e.g., Crowder, 1982), but Baddeley and his colleagues (Baddeley & Hitch, 1974) revitalized the short-term/long-term distinc- tion in the mid 1970s. They changed the direction of research on short-term memory by asking whether the memory system used in typical short-term memory tasks is a ‘‘working memory’’ system (i.e., a memory system that is used for the temporary storage and manipulation of information in the service of complex tasks). Because tasks such as reasoning and reading were impaired when short- term memory was filled to capacity by a concurrent task, these investigators concluded that short-term memory tasks do indeed tap into a working memory system. Moreover, they developed a model of working memory in which modality-specific slave systems are used for storing information, and a central executive is used to read, write, and manipulate this information. In this manner, Badde- ley and his colleagues ‘‘rescued’’ the concept of short-term memory, demonstrating that it can be distinguished from long-term memory and that it plays an important role as a working memory system (Baddeley & Hitch, 1974). r 2003 by the Society for Research in Child Development, Inc. All rights reserved. 0009-3920/2003/7406-0015 Shannon Ross-Sheehy, Lisa M. Oakes, and Steven J. Luck, Department of Psychology, University of Iowa. A portion of these data were presented at the biennial meeting of the Society for Research in Child Development, 2001, and the biennial meeting of the International Society for Infant Develop- ment, 2002. This research and preparation of this manuscript were made possible by a University of Iowa Obermann Center for Advanced Studies Spelman Rockefeller grant, and by National Institutes of Health Grants R01 HD36060, R03 MH64020, R01 MH63001, and R01 MH65034. We would like to thank the undergraduates at the University of Iowa Infant Perception and Cognition lab for their help with the data collection and coding phases of this project. Correspondence concerning this article should be addressed to Lisa M. Oakes, Department of Psychology, University of Iowa, Iowa City, IA 52242. Electronic mail may be sent to lisa- oakes@uiowa.edu. Child Development, November/December 2003, Volume 74, Number 6, Pages 1807– 1822