Working memory capacity predicts the beneficial effect of selective memory retrieval

Abstract

Selective retrieval of some studied items can both impair and improve recall of the other items. This study examined the role of working memory capacity (WMC) for the two effects of memory retrieval. Participants studied an item list consisting of predefined target and nontarget items. After study of the list, half of the participants performed an imagination task supposed to induce a change in mental context, whereas the other half performed a counting task which does not induce such context change. Following presentation of a second list, memory for the original list's target items was tested, either with or without preceding retrieval of the list's nontarget items. Consistent with previous work, preceding nontarget retrieval impaired target recall in the absence of the context change, but improved target recall in its presence. In particular, there was a positive relationship between WMC and the beneficial, but not the detrimental effect of memory retrieval. On the basis of the view that the beneficial effect of memory retrieval reflects context-reactivation processes, the results indicate that individuals with higher WMC are better able to capitalise on retrieval-induced context reactivation than individuals with lower WMC.

The two faces of selective memory retrieval: Earlier decline of the beneficial than the detrimental effect with older age.

“…In this experiment, the beneficial effect of selective retrieval was reduced in older adults, but we found no evidence for a role of WMC for this age-related reduction. At first glance, this finding seems to be at odds with prior work (Schlichting et al, 2015) and to suggest that age-related working memory decline cannot account for the present age-related decline in the beneficial effect. However, in the prior work, the complex operation span task was employed to measure participants' WMC, a task that may be a more sensitive measure of WMC than the simple span task used in this experiment (e.g., Conway et al, 2005).…”

“…Because low-WMC individuals are often more vulnerable to blocking effects than high-WMC individuals (e.g., Aslan & Bäuml, 2010; Kane & Engle, 2000), the typically positive relationship between WMC and inhibitory efficiency (Aslan & Bäuml, 2011; Redick, Heitz, & Engle, 2007) may be masked by the induced blocking. Such reasoning can explain why a positive relationship between WMC and retrieval-induced forgetting was present in studies in which target recall was delayed (e.g., Aslan & Bäuml, 2011) but was absent in studies in which no such delay occurred (e.g., Schlichting et al, 2015, present Experiment 2).…”

“…Recent work provided evidence for a critical role of WMC for the beneficial effect of selective retrieval in young adults (Schlichting et al, 2015). In this study, a positive relationship between the beneficial effect and WMC was found, suggesting that low-WMC individuals are less able than high-WMC individuals to benefit from retrieval-induced context reactivation.…”

“…Following the view that the beneficial effect of retrieval is mediated by context-reactivation processes, and recent (indirect) evidence that older adults may have difficulty capitalizing on such processes (Kahana et al, 2002), we expected a beneficial effect of selective retrieval in younger adults, but a reduced or even eliminated effect in older adults. Finally, on the basis of the recent results on the role of WMC for the beneficial effect of retrieval (Schlichting et al, 2015), we expected the age-related decline in the beneficial effect to be related to the age-related decline in WMC.…”

Part-list cuing effects in younger and older adults’ episodic recall.

“…Rather, due to their deficits in self-initiated processing (Craik, 1986; Craik & Byrd, 1982), older adults may be less likely than younger adults to spontaneously initiate and engage in context-reactivation processes when provided with part-list cues. Although tentative at this point, this notion would be consistent with previous research indicating that (self-initiated) context reactivation is a resource-demanding process that depends on working memory capacity (Schlichting, Aslan, Holterman, & Bäuml, 2015; Spillers & Unsworth, 2011), and that working memory capacity is reduced in older age (Craik & Salthouse, 2008).…”