The Development of Clustering in Episodic Memory: A Cognitive‐Modeling Approach

Abstract

Younger children’s free recall from episodic memory is typically less organized than recall by older children. To investigate if and how repeated learning opportunities help children use organizational strategies that improve recall, the authors analyzed category clustering across four study‐test cycles. Seven‐year‐olds, 10‐year‐olds, and young adults (N = 150) studied categorically related words for a free‐recall task. The cognitive processes underlying recall and clustering were measured with a multinomial model. The modeling revealed that developmental differences emerged particularly in the rate of learning to encode words as categorical clusters. The learning curves showed a common pattern across age groups, indicating developmental invariance. Memory for individual items also contributed to developmental differences and was the only factor driving 7‐year‐olds’ moderate improvements in recall.

When judging what you know changes what you really know: Soliciting metamemory judgments reactively enhances children’s learning

“…Because it is well‐known that young children are less able to sustain their attention and maintain their effort in a learning or cognitive task than older children (Klenberg et al, 2001; Ruff et al, 1990; Wetzel, 2014), there should be more opportunity for the requirement to make concurrent JOLs to enhance young children's study effort and to boost their memory performance. Equally, because young children are less able to employ effective (complex) encoding strategies when they are not explicitly or implicitly prompted to do so (Horn et al, 2021; Schleepen & Jonkman, 2012; Schneider, 1986), there should be more opportunity for concurrent JOLs to refine their learning strategies. Therefore, the positive‐reactivity theory predicts a negative interaction between age and reactivity, with younger children benefiting more from making concurrent JOLs than their older counterparts.…”

The Development of Attention to Objects and Scenes: From Object‐Biased to Unbiased

“…This is in contrast to standard statistical models, such as analysis of variance (ANOVA) or regression models, which are atheoretical and do not address mechanisms contributing to performance without additional inferences on the part of the researcher (Lewandowsky & Farrell, 2011). MPT models have been productively used in prior developmental work to estimate the contribution of cognitive processes such as memory binding and semantic clustering to task performance (Horn, Bayen, & Michalkiewicz, 2020; Yim, Dennis, & Sloutsky, 2013). The MPT model we developed for the current work parameterized attention to relevant as well as irrelevant streams of information, allowing us to gain insight into developmental differences in focusing and filtering of scene and object information.…”

The underlying processes of episodic memory development: From a unique contribution of representation to the increasing use of semantic organization supported by cognitive control

“…Relatedly, although the ARC measure is a reliable proxy of clustering, it nevertheless merges both encoding, storage and retrieval processes into one measure. Therefore, other methods such as a cognitive modelling approach could be used in the future to better disentangle the relative contributions of Representation and Control to encoding and retrieval underlying recall and clustering (see Horn et al, 2021 for a cognitive modelling approach on clustering in children' episodic memory). Another limitation relates to the use of interindividual comparisons instead of intra-individual comparisons when looking at organizational behaviors, and the former potentially creates more utilization deficiencies than the latter and this can lead to misleading conclusions about age group comparisons of utilization deficiency (Schlagmüller & Schneider, 2002).…”

“…Another limitation relates to the use of interindividual comparisons instead of intra-individual comparisons when looking at organizational behaviors, and the former potentially creates more utilization deficiencies than the latter and this can lead to misleading conclusions about age group comparisons of utilization deficiency (Schlagmüller & Schneider, 2002). As the literature is currently mixed between using one or the other approach (e.g., Horn et al, 2021;Miotto et al, 2020;Schleepen & Jonkman, 2011), a potential future study should investigate to what extent the use of these two approaches does influence the observed results regarding organizational behaviors, and more especially utilization deficiency and which types (Bjorklund et al, 1997). Finally, we acknowledge that although our sample size was large enough for regression analyses (see Participants section), it was relatively small for correlational analyses as Schönbrodt and Perugini (2013) have demonstrated that correlation coefficients tend to stabilize with a sample size of around 250 participants.…”

“…When presented with semantically related items (lexical sets, e.g., 'glass', 'cup', 'saucer'), although it has been first reported that children aged 9 to 10 years show some evidence of clustering, organizing their responses by recalling items in semantic or adjacent groups (Hasselhorn, 1990) , more recent studies showed that this ability progressively develops from 7 to 13 years-old (e.g., Bjorklund & de Marchena, 1984;Bjorklund & Jacobs, 1985;Schleepen & Jonkman, 2012;Schwenck et al, 2007; for a review see Ornstein et al, 2010) . Indeed, a recent cognitive modelling study has shown that memory for individual items is the only factor contributing to enhance memory performance in 7-year-old children whereas encoding items as clusters increasingly predicts better performance for children older than 10 years-old (Horn et al, 2021).…”