Retrieval practice can insulate items against intralist interference: Evidence from the list-length effect, output interference, and retrieval-induced forgetting.

Abstract

This study sought to determine whether nonselective retrieval practice after study can reduce memories' susceptibility to intralist interference, as it is observed in the list-length effect, output interference, and retrieval-induced forgetting. Across 3 experiments, we compared the effects of nonselective retrieval practice and restudy on previously studied material with regard to these 3 forms of episodic forgetting. When study of an item list was followed by a restudy cycle, recall from a longer list was worse than recall from a shorter list (list-length effect), preceding recall of studied nontarget items impaired recall of the list's target items (output interference), and repeated selective retrieval of some list items attenuated recall of other nonretrieved items at test (retrieval-induced forgetting). In contrast, none of these effects arose when study of the list was followed by a nonselective retrieval cycle. The findings are consistent with a combination of contextual variability theory and a variant of study-phase retrieval theory that assumes that retrieval can create more distinct context features for retrieved items than restudy does for restudied items, thus reducing items' susceptibility to interference relative to restudy cycles. The findings add to the view that nonselective retrieval practice can stabilize and consolidate memories.

Reversing the testing effect by feedback: Behavioral and electrophysiological evidence

“…According to this modification, only the not-recalled items, but not the recalled items, are strengthened by feedback, regardless of practice condition. With regard to the cognitive processes that may underlie the strengthening of items, and following the suggestion that alpha/lower-beta power decreases during item encoding index semantic or deep item encoding (Hanslmayr et al, 2009;Klimesch et al, 1997), the present results are particularly consistent with the semantic elaboration account of testing and test-potentiated learning effects (Carpenter, 2009;Pyc & Rawson, 2010), although other results clearly indicate an additional role of unique context cues at test, as it is suggested by the episodic context account of retrieval-based learning (e.g., Kliegl & Bäuml 2016;Lehman et al, 2014). The bifurcation model is well compatible with the idea that both semantic elaboration and context updating via retrieval practice can contribute to memory strengthening of (the previously not-recalled) items.…”

“…Context updating via retrieval practice thus enhances memory search and improves target recall. The episodic context account can provide explanations for a number of findings in the retrieval-based learning literature (e.g., Kliegl & Bäuml 2016;Lehman, Smith, & Karpicke, 2014).…”

“…In addition to the testing effect, other benefits of retrieval practice can arise. For instance, retrieval practice in comparison to restudy practice can enhance transfer of learning (Butler, 2010;Carpenter & Kelly, 2012), potentiate new learning Pastötter, Schicker, Niedernhuber, & Bäuml, 2011), and insulate tested material against interference from other information (Halamish & Bjork, 2011;Kliegl & Bäuml, 2016). On the basis of these and related findings, it has been argued that retrieval practice is a more effective learning strategy than restudy practice, in both the laboratory and the classroom (Dunlosky, Rawson, Marsh, Nathan, & Willingham, 2013;Roediger & Butler, 2011).…”

Does retrieving a memory insulate it against memory inhibition? A retroactive interference study

“…Consistent with this view, recent findings suggest that retrieval can, indeed, protect retrieved items from retroactive interference arising from novel encoding (Halamish & Bjork, 2011), proactive interference from prior lists (Pastötter, Schicker, Niedernhuber, & Bäuml, 2011), and directed forgetting (Abel & Bäuml, 2016). Critically, several studies have suggested that an initial retrieval attempt may insulate a memory against the inhibitory processes thought to create retrieval-induced forgetting (Kliegl & Bäuml, 2016;Racsmány & Keresztes, 2015). We put this finding to the test using a classical A-B, A-C retroactive interference design in the current experiments.…”

“…Recently, however, several findings have suggested an intriguing exception to this tendency: Boosting the accessibility of a memory through an initial round of non-selective retrieval practice can eliminate retrievalinduced forgetting. For example, Kliegl and Bäuml (2016) found that subjecting recently encoded items to a single cycle of non-selective retrieval practice rendered those items resilient to later retrieval manipulations that would usually induce forgetting. By modifying the standard retrieval-practice paradigm, they demonstrated that retrievalinduced forgetting of category-cued exemplars fails to emerge when the selective retrieval phase is preceded by an opportunity to recall all of the previously studied items.…”