4.1.2 - Explanations for forgetting

Availability and accessibility

AQA wants two explanations for forgetting in this lesson: interference and retrieval failure due to absence of cues. Both are explanations of why a person may fail to recall information, especially from long-term memory. They do not simply say "time passed, so the memory disappeared." They explain what makes recall difficult.

A useful first distinction is availability versus accessibility:

Interference usually explains forgetting as competition between memories. Retrieval failure usually explains forgetting as poor access to a stored memory. In real life they can overlap: a cue might be missing, and similar memories might also compete. In AQA answers, the skill is to identify which explanation is doing the most work in the scenario.

AO1 requires accurate definitions and named types. AO2 requires applying the direction of forgetting to a person or situation. AO3 requires judging how well the explanation is supported by research and where it is limited.

Proactive and retroactive interference

Interference is most likely when two memories are similar. Similar memories can share the same retrieval cue, so the cue points to more than one possible answer. This creates response competition: the person retrieves the wrong memory, mixes memories, or cannot choose confidently between them.

Use the direction of learning to diagnose the type:

The direction matters because AQA often gives a short scenario and expects precise application. If a student says only "interference happened", they may miss marks that depend on naming the correct type.

Similarity is not a minor detail; it is central. Learning French after Spanish is more likely to create interference than learning French after a topic with fewer shared cues, such as trigonometry. French and Spanish vocabulary can both be cued by "foreign language word for house", so the wrong candidate memory may be retrieved.

For example, Aisha learned a monologue for a school performance last year and a different monologue this year. If she is asked to perform last year's piece and lines from this year's piece intrude, that is retroactive interference: the new learning disrupts the old. If she tries to perform this year's piece but keeps saying last year's lines, that is proactive interference: the old learning disrupts the new.

Evidence for interference

Classic laboratory research supports the idea that similarity creates interference. McGeoch and McDonald (1931) investigated retroactive interference by asking participants to learn an original list of adjectives, then learn a second list that varied in similarity to the first. Some second lists were closely related, such as synonyms, while others were less similar. Recall of the original list was worse when the second list was more similar.

The conclusion is that later learning can disrupt earlier learning, especially when the memories are semantically similar. This supports retroactive interference and the idea of response competition: when the cue is too broad or points to several similar memories, the wrong one may be retrieved.

Another useful study is Baddeley and Hitch (1977), who studied rugby players' memory for teams they had played. Some players had missed games, so the researchers could separate "time since the match" from "number of intervening matches". Players who had played more intervening games tended to recall fewer earlier teams. This supports interference in a real-life setting because recall was related to intervening similar events rather than simply elapsed time. However, missed matches were not randomly assigned: injury, selection and how much attention a player paid could also differ. The pattern supports interference but is weaker causal evidence than a controlled manipulation of similarity.

Judging the evidence

Good exam evaluation should avoid vague phrases such as "low ecological validity" by itself. Say what the task was, why that task differs from everyday memory, and how that affects the explanation. For example, "word-list recall may exaggerate interference because the experiment deliberately creates similar material close together, whereas many everyday memories have richer, more distinctive cues."

Retrieval failure and cues

Retrieval failure says forgetting can happen even when the memory is still stored. The problem is that the person lacks the right cue to access it.

The key principle is encoding specificity, associated with Tulving and Thomson (1973). A cue is most useful when it was present, or meaningfully connected with the material, at the time of encoding and is also available at retrieval. In plainer language: a cue helps when it matches how the memory was originally stored.

Two AQA-friendly forms are:

Context-dependent forgetting occurs when external cues present at learning are absent at recall. A student might remember more when sitting at the same desk where they revised because the desk, room, lighting, and routine act as cues. If those cues are absent, the memory may be less accessible.

State-dependent forgetting occurs when internal cues present at learning are absent at recall. A person who learned information while drowsy, stressed, or in a particular mood may find recall easier when in a similar internal state. This does not mean students should try to recreate harmful or impractical states. The exam point is that internal states can act as retrieval cues.

Retrieval failure is especially useful because it explains the common "tip-of-the-tongue" experience: the information feels stored, but the cue is not strong enough. A later prompt, category heading, first letter, place, or smell can make the memory accessible again.

Evidence for retrieval failure

Tulving and Pearlstone (1966) found that category cues could improve recall of word lists. This supports the distinction between availability and accessibility: information not recalled freely may still be available if a cue makes it accessible.

Godden and Baddeley (1975) tested context-dependent memory using divers. Participants learned word lists either on dry land or underwater and then recalled them in the same or a different environment. Recall was better when the learning and recall contexts matched. This supports retrieval failure because changing the external context removed cues that had been present during encoding.

Carter and Cassaday (1998) tested state-dependent retrieval using an antihistamine, chlorpheniramine, and a placebo. University students learned material and recalled it later either in the same drug state or a different drug state. Recall was better in congruent conditions, such as antihistamine at learning and antihistamine at recall, than in incongruent conditions. This supports the claim that internal state can act as a retrieval cue.

Interpreting a matching-context pattern

Here are original illustrative mean scores out of 20, not Godden and Baddeley’s reported values:

Both matching conditions outperform their corresponding mismatching condition. Mean recall is $(16+15)/2 = 15.5$ when contexts match and $(10+9)/2 = 9.5$ when they differ, a difference of 6 words. Room A is not simply always better: after learning in B, recall is better in B. That pattern supports a match between encoding and retrieval cues. Equivalent lists and balanced condition order are needed so list difficulty or practice does not provide the explanation.

Judging the evidence

Retrieval failure is a strong explanation for temporary forgetting and cue-dependent recall. Its weakness is that it does not explain all forgetting. Sometimes the original encoding was poor, the memory trace may have changed, or similar memories may compete. A balanced answer can say retrieval failure is powerful for explaining why cues improve recall, but it is not a complete account of every kind of forgetting.

Distinguishing competing explanations

Different patterns of forgetting allow us to compare the explanations, but one observation rarely rules out every alternative.

A single failure to recall cannot establish its cause. Both interference and missing cues can leave a memory temporarily inaccessible. Discriminating evidence is a result predicted more clearly by one explanation than by the other.

For interference, compare recall of the same original material after similar versus dissimilar new learning. Hold the delay, list length and learning time constant. Worse recall after similar material supports competition; a change in delay as well as similarity would make the cause ambiguous. Intrusions from the new list into recall of the old list specifically support retroactive interference.

For retrieval failure, compare matching and mismatching learning/recall contexts with the same material and no extra competing learning. Better recall when an encoding cue returns supports cue-dependent access. It does not show that all forgetting has this cause. A person may benefit from a cue because it distinguishes two competing memories, so the mechanisms can overlap.

Consider a learner who studies two similar sets of vocabulary and later remembers more when a topic heading is supplied. Similarity supports an interference account; improvement with a heading supports a retrieval account. Both may operate, and the heading may help select the target list. A justified interpretation states what changed, what was measured, and which alternatives remain.

This distinction also guides evaluation. Controlled word-list research can isolate a mechanism but may exaggerate competition compared with rich everyday memories. Field research can show relevance outside the laboratory but has less control over initial learning, opportunities for rehearsal and distinctiveness of cues.

Quick recap

Proactive interference means older learning disrupts newer learning; retroactive interference reverses that direction. Similarity can create competition. Retrieval failure concerns available information that lacks effective retrieval cues; matching external contexts or internal states may help. Both mechanisms can affect accessibility, and neither explains every instance of forgetting.