Retrieval-induced forgetting

Learn several fruits and several tools. Practicing only some fruit names may make the other fruit names harder to recall than the unpracticed tools. This is retrieval-induced forgetting. Has practice made the competing fruits less accessible, or changed which names your cues bring to mind?[1]

Suppose you learn fruit and tool names, then practice recalling only some fruits. Later, the unpracticed fruits may be harder to recall than the unpracticed tools. This comparison defines retrieval-induced forgetting. What changes: the accessibility of competing memories, or the cues and competition that determine which memory is retrieved?

An inhibition account proposes that selective retrieval reduces a competitor’s accessibility. A context or competition account can produce a recall cost because practiced items are favored by the retrieval situation, without destroying the stored content. The comparison with an entirely unpracticed category helps isolate the selective cost rather than ordinary forgetting.

The cited 2026 study of young adults found the cost on successive tests, with a smaller difference on the second. Its exercise manipulation did not preserve the cost as predicted; it did not establish a physiological cause of forgetting.[1]

Was access to the competing fruit names inhibited, or did practice make other names and cues dominate? When the cost shrinks on a later test, time and the earlier test can both be responsible.[1]

The second test followed both extra time and an earlier retrieval opportunity, so the study cannot determine which caused attenuation. Equal-delay groups with and without that earlier test would provide the missing contrast.

There is also a distinction between losing an item and losing one route to it. As an invented example, “yellow fruit” might bring back a banana that the practiced “fruit” retrieval cue failed to retrieve. If such a change were demonstrated under controlled cues, it would constrain an access explanation; one successful or failed cue would still not by itself prove inhibition or memory erasure.[1]

The distinction prevents a recall failure from being mistaken for proof that a memory was destroyed.[1]

Selective remembering changes which parts of learned knowledge are available next. This makes retrieval-induced forgetting a connection between learning and access, rather than simply a count of facts retained: improving recall of one item can coincide with a cost for its neighbors.

The distinction matters when interpreting later recovery. If the retrieval situation has changed, an item’s return need not mean a destroyed memory was rebuilt. If prior testing contributed to recovery, the test was part of the learning history rather than a passive measurement. Explaining these possibilities would clarify what a recall failure tells us about stored knowledge, with no general claim that retrieval practice is harmful.[1]