Can remembering one thing make you forget another? Retrieval-induced forgetting
The idea
Practising a memory makes it easier to recall later. That part is unsurprising. The less obvious claim is that practising some memories can make related, unpractised memories harder to recall. Psychologists call this retrieval-induced forgettingretrieval-induced forgetting The finding that practising recall of some memories can make related memories that were not practised harder to recall later..
The standard demonstration comes from Michael Anderson, Robert Bjork and Elizabeth Bjork, published in 1994. In their first experiment, 36 students studied category and example pairs, such as a fruit paired with “orange”: six examples from each of eight categories. In a practice phase they repeatedly recalled half of the examples from half of the categories, prompted with the category name and the first letters of the word. After a 20-minute gap came a surprise test asking for everything they could remember from each category.
That design produces three kinds of item. Practised items (labelled Rp+) were recalled 73.6 percent of the time. Items from categories that were never practised (Nrp) serve as the baseline, at 48.4 percent. The interesting group is the unpractised items from practised categories (Rp-): these were recalled 37.5 percent of the time, worse than the baseline. Practising “orange” seemed to hurt memory for “banana”, but not for items in unrelated categories.
The authors also reported that the impairment was concentrated in strong, common category members rather than weak, rare ones. Their interpretation was that when you try to retrieve one item, other strong candidates compete for recall, and the memory system suppresses them. That suppression account, usually called inhibition, became the dominant explanation.
Why it might change a decision
The finding bears on any situation in which some details are rehearsed and related details are not: a revision session that covers half of a topic, or an interview in which a witness is asked repeatedly about some aspects of an event and not others. The testable prediction is specific. After selective practice, recall of the unpractised but related material should be lower, at a short delay, than recall of comparable material from a topic that was not practised at all. Someone running a training course could test this directly by quizzing on half of a module and comparing later recall of the other half with an unquizzed module.
What the laboratory work does not establish is how large or how lasting the effect is in such settings. The experiments mostly use word lists and delays of minutes, and Rowland and colleagues list long retention intervals among the conditions under which the effect has been reported to fail.
How solid is this?
The core result does not rest on a single study. Christopher Rowland, Lauren Bates and Edward DeLosh ran a set of replications published in 2014. In their first experiment, 72 undergraduates (a sample sized by a power analysis) showed the standard pattern: unpractised items from practised categories were recalled at 41 percent against 45 percent for baseline items, an effect sizeeffect size How large an effect is, as opposed to whether it exists at all. A result can be statistically reliable and still be too small to matter. of d = 0.31. Practised items were recalled at 79 percent. The effect was also found in an online sample. Note how much smaller the forgetting gap was than in 1994: about four points rather than eleven.
The explanation is less settled than the effect. The strongest competing account is interference: strengthening the practised items may simply make them come to mind first at test and block their neighbours, without any active suppression. One way to separate the accounts is an independent-cue testindependent-cue test A memory test that asks for an item using a different prompt from the one it was learned with, to see whether forgetting follows the item itself., which probes an unpractised item through a different category (asking for “banana” with the cue “monkey” rather than “fruit”, say). Inhibition predicts the forgetting should follow the item to the new cue; blocking predicts it should not. Rowland and colleagues did not find retrieval-induced forgetting with the independent-cue procedure. Their introduction also notes that failures to replicatereplication Running a study again, independently, to see whether the original result holds up. under conditions close to the original have been reported, and that such failures are rarely published, which is a reason to treat the published record as possibly inflated.
The broadest summary is a meta-analysismeta-analysis A study that pools the results of many earlier studies on the same question to estimate an overall effect and what makes it vary. by Kou Murayama, Toshiya Miyatsu, Dorothy Buchli and Benjamin Storm in Psychological Bulletin, described by its authors as the first major meta-analysis of the effect. Their abstract says the results largely supported inhibition accounts but also provided some challenging evidence, and that the results often varied depending on how the forgetting was measured. In plain terms: the effect is real enough to survive pooling, but whether one mechanism explains it, and how much it depends on test format, remains contested.
Popular accounts sometimes present the effect as a general rule that recalling anything erases its neighbours. The data support something narrower: a modest, reasonably reliable impairment under specific laboratory conditions, with its reach into everyday memory still an open question.
Sources
- Anderson, M. C., Bjork, R. A., and Bjork, E. L. (1994). Remembering can cause forgetting: Retrieval dynamics in long-term memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 20(5), 1063-1087. PDF, Bjork Learning and Forgetting Lab
- Rowland, C. A., Bates, L. E., and DeLosh, E. L. (2014). On the reliability of retrieval-induced forgetting. Frontiers in Psychology, 5, 1343. doi:10.3389/fpsyg.2014.01343
- Murayama, K., Miyatsu, T., Buchli, D., and Storm, B. C. (2014). Forgetting as a consequence of retrieval: A meta-analytic review of retrieval-induced forgetting. Psychological Bulletin, 140(5), 1383-1409. doi:10.1037/a0037505