Electrophysiological evidence for late maturation of strategic episodic retrieval processes

Abstract

Improvement in source memory performance throughout development is thought to be mediated by strategic processes that facilitate the retrieval of task-relevant information. Using event-related potentials (ERPs), we examined developmental changes in these processes during adolescence. Adolescents (13-14 years) and adults (19-29 years) completed a memory exclusion task which required the discrimination between words studied in one color ('targets') and words studied in the alternative color ('non-targets') under two conditions that put different demands on strategic control. Memory accuracy improved with age and also increased with decreasing control demands in both age groups. The parietal old/new effect, an ERP correlate of recollection, was reliable for targets across conditions in both age groups. By contrast, ERP correlates of non-target recollection were present in adolescents across conditions but not in adults. This suggests that adults implemented a strategy to prioritize recollection of target information with greater success than adolescents regardless of control demands, presumably reflecting maturational differences in cognitive control. In support of this view, the ERP amplitude difference between targets and non-targets was positively correlated with a measure of working memory capacity (WMC) in adults but not in adolescents. A further age-related difference was that ERP correlates of post-retrieval processing, including late right-frontal old/new effects and late posterior negativities, were observed in adults only. Together, our data suggest protracted maturation in the strategic processes that underlie selective recollection and post-retrieval control.

Aging, working memory capacity and the proactive control of recollection: An event-related potential study

“…The early old/new effect was present only for non-targets in the young, but for both conditions in the older group. In their developmental study, Sprondel et al [33] observed an unexpected frontal positive-going old/new effect in their young adults which shared some features with the present findings, being present only for non-targets, although it occurred after the mid-frontal effect, overlapping temporally with their LP effect (500–700 ms). Finding that this effect was larger in higher-WMC young adults, the authors proposed that it reflected relatively early cue-specification processes used to discriminate between targets and non-targets.…”

“…We also expected that in young adults, recollection selectivity would be greater in individuals with greater WMC, replicating Elward, Wilding and others’ findings [28,30]. Critically, we predicted that the relation between WMC and recollection selectivity would not be observed in the older adults, converging with findings from other populations with reduced attentional control [30,33], and supporting the proposal that older adults fail to engage attentional resources in order to control recollection. This would also provide further support for the involvement of proactive attentional control in pre-retrieval control of episodic memory.…”

“…However, there is converging ERP evidence linking developmental changes in cognitive control to pre-retrieval processing at the other end of the lifespan. A reduced difference between target and non-target left parietal old/new effects has also been shown in adolescents relative to young adults [33], analogous to Dywan et al’s finding in older adults. Furthermore, individual differences in WMC correlated with this ERP difference measure only in the young adults, who may also have differed from adolescents in this regard.…”

“…However, although the multiple regression model (with the same predictors as for the LP effect) was significant overall ( F (4,30) = 4.6, MSE = 3.4, p = .005), this was driven by an intercept, i.e., a significant mean old/new effect; none of the predictors of interest were significant. Because of Sprondel et al’s [33] prior finding of a positive relation between a frontal non-target old/new effect magnitude and O-span in young adults (but not adolescents), we also checked for a similar effect, but found no hint of any significant association in the present young group ( F (1,18) = .03,MSE = 2.5, p = .87).…”

How does testing affect retrieval-related processes? An event-related potential (ERP) study on the short-term effects of repeated retrieval

“…However, this apparently had no impact on the LPN. There is also evidence from other studies that task difficulty (or the efficiency with which task-relevant attribute conjunctions can be recovered) does not necessarily affect the LPN amplitude (Sprondel, Kipp, & Mecklinger, 2012).…”