The developmental progression in short-term memory (STM) is well-documented throughout childhood. Most research has emphasized the verbal memory domain, and less is known about the visual-spatial systematic improvement. We conducted a cross-sectional study assessing 4 age groups of preschool and school-aged children (N = 223): Age Group 1 (mean age = 50 months), Age Group 2 (mean age = 71 months), Age Group 3 (mean age = 87 months) and Age Group 4 (mean age = 96 months). STM performances across all age groups, as well as preschoolers' metamemory judgments (concerning their visual-spatial memory strategies) were investigated. Regarding response accuracy, we found that span performance increased across the four age groups on all span tasks. School aged children made a better use of verbal recoding of familiar visual stimuli (colors and objects); only on these visual, but verbally-recodable tasks did we also find a gender effect in preschool children, girls outperforming boys. The total number of errors on verbal STM decreased with age. Finally, regarding metamemory, most preschoolers were able to make judgments regarding the strategies they used to better encode the memoranda (i.e. spontaneous rehearsal and verbal coding of visual stimuli).
KEYWORDS: short-term memory, simple span, children, error analysis, metamemory.
Short-term memory (STM) represents the temporarily increased availability of information in memory that can be used to carry out various types of mental tasks (Cowan et al., 1999). The STM store is limited in its capacity: although through the use of mnemonic strategies (e.g. chunking) adults can remember up to seven elements (Miller, 1956), the real capacity limit is of about three or four elements, similar to the amount that can be simultaneously sustained in the focus of attention (Cowan, 1995; Halberda, Simons, & Wetherhold, 2006), or to what Sperling (1960) called "the span of apprehension". The story of this construct has been inextricably linked to the one of working memory (WM), the latter representing an active memory system that is responsible for the temporary maintenance and simultaneous processing of information that is typically required in complex cognitive tasks (Baddeley, 1986).
A note should be made at this point: in infants and very young children, the distinction between WM and STM is blurred and the perspectives are contradictory; we will present the main arguments of the existing positions since the basic reasoning can be generalized to older preschoolers. On one hand, researchers such as Reznick (2007) argue that short-term retention in infancy is actually an index of WM. He concludes this after a critical analysis of the most widely used testing procedures of short-term retention in infants: the hide-find method (a delayed-response task), observe-perform procedures (action sequence re-enactment after a delay), and the familiarize-recognize procedure. The critical analysis leads to a common caution regarding all the tasks related to the ability of the research design to rule out possible confounds (alternative explanations): long-term memory mechanisms, reactions to violations of expectations, motivation or motor competence. This researcher's preference for the term working memory (although not identical to the significance of the term in adults) in order to characterize very young children's short-term retention is based on three motivations: 1) the "naturalistic" request for WM, as opposed to the more artificial STM tasks, 2) the executive processing requirements, specific to WM tasks; 3) the existing proofs for the very early emergence of short-term retention in young children during problem-solving scenarios. On the other hand, researchers such as Oakes, Ross Sheehy, and Luck (2007) make the opposite claim, suggesting that it is difficult to assess the "workings" of WM in infants and very young children, because in their case the processes might be more bottom-up driven, thus favoring the alternative short-term memory terminology for characterizing their memory processes. …