The origin of the interaction between learning method and delay in the testing effect: The roles of processing and conceptual retrieval organization

Abstract

Recent research has demonstrated a relationship between retrieval organization and the efficacy of prior repeated retrieval on delayed tests. The present study asked why repeated study engenders higher recall at a short delay despite lower retrieval organization but produces a decline at a long delay, and why repeated retrieval engenders lower recall at a short delay despite higher retrieval organization but produces stable recall over time. This relationship was examined through the inclusion of two successive recall tests-one immediately after learning method and one a week later. Results replicated the interaction in recall between learning method and delay characterizing the testing effect and, critically, revealed the qualitative differences inherent in the retrieval organization of each method. Specifically, stable recall in repeated retrieval was accompanied by strong and sustained conceptual organization, whereas organization for repeated study was tenuous and weakened across tests. These differences quantitatively were assessed through the use of five targeted analyses: specifically, the examination of cumulative recall curves, the accumulation of organization across time (a curve akin to cumulative recall), item gains and losses across time, changes in the size of categories across time, and the fate of specific clusters of recalled items across time. These differences are discussed within the context of differential processes occurring during learning method.

Comparing the testing effect under blocked and mixed practice: The mnemonic benefits of retrieval practice are not affected by practice format

“…This retrieval-specific boost in final-test performance was similar in size after 5 min and 7 days. While some previous studies found the testing effect to increase or only to occur with long retention intervals when tests did not provide feedback (e.g., Bäuml, Holterman, & Abel, 2014;Congleton & Rajaram, 2012; 2006), the present experiment is consistent with other reports of similar effects after a short and a long delay when feedback is provided on the initial tests (e.g., Carpenter et al, 2008). The data reported by Kornell, Bjork and Garcia (2011, Exp. 2), moreover, support the idea that provision of feedback after retrieval attempts may be a crucial difference between studies that show immediate testing effects and those that do not (like Roediger & Karpicke, 2006, and others).…”

Retrieval practice does not safeguard memories from interference-based forgetting

“…The present study assessed the competing hypotheses that testing returns memories to a vulnerable state in which they are especially susceptible to RI and intrusions as suggested by the reconsolidation account, or that testing protects memories from the potential negative effects of subsequently learned material as suggested by the list-separation hypothesis (e.g., Congleton & Rajaram, 2012;Halamish & Bjork, 2011;Szpunar et al, 2008). Before evaluating the hypotheses, it is important to note that our paradigm was generally sensitive to the testing effect: testing List 1 in Session 2 enhanced List 1 memory in comparison to the subtle reminder and the no reminder condition.…”

“…Based on the idea that testing promotes list separation and enhances list discriminability (e.g., Congleton & Rajaram, 2012;Szpunar et al, 2008), testing List 1 memory before learning the second list of objects should reduce intrusions from List 2 into List 1. Additionally, based on Potts and Shanks (2012), we should see reduced RI in the form of enhanced memory in the testing condition in comparison to a subtle reminder or no reminder condition.…”

“…It uses a reconsolidation paradigm (Hupbach, Gomez, Hardt, & Nadel, 2007;Hupbach, Hardt, Gomez, & Nadel, 2008;Hupbach, Gomez, & Nadel, 2009;Hupbach, Gomez, & Nadel, 2011) to assess how testing in comparison to other reactivation procedures affects declarative memory. This paradigm will allow for a direct evaluation of the hypotheses that testing makes declarative memories vulnerable to interference, or that testing protects memories from the potential negative effects of subsequently learned material, as suggested by the list-separation hypothesis (e.g., Congleton & Rajaram, 2012;Halamish & Bjork, 2011;Szpunar et al, 2008). This question has important practical implications.…”

Effects of successive relearning on recall: Does relearning override the effects of initial learning criterion?

“…Bjork & Bjork, 1992;Loftus, 1985). Recent evidence also suggests that testing can slow forgetting (e.g., Congleton & Rajaram, 2012;Wheeler, Ewers, & Buonanno, 2003). If testing does slow forgetting, then the rates of forgetting might differ, given differences in the initial learning criterion.…”