Explanations for Forgetting: Interference and Retrieval Failure

AQA requires you to know two explanations for forgetting: interference and retrieval failure due to the absence of cues. Both mainly explain forgetting in long-term memory (LTM), where information is stored over long periods but sometimes cannot be recalled.

The two explanations disagree about why a memory is lost:

Explanation	Core idea	Is the memory still there?
Interference	Two memories disrupt each other, so one blocks recall of the other	The memory may be lost or blocked
Retrieval failure	The right cue is missing at recall	Yes — it is stored but not accessible

Interference is about two competing memories getting in each other's way. Retrieval failure is about a missing trigger for a memory that is still stored.

Keeping these two mechanisms separate is the foundation for the whole topic, because exam questions frequently ask you to distinguish them or to apply the correct one to a scenario.

Interference occurs when two pieces of information compete with each other, making it harder to recall one or both. There are two directions this can take.

• Proactive interference (PI): old information disrupts the learning or recall of new information. For example, your old phone number keeps coming to mind when you try to remember your new one.
• Retroactive interference (RI): new information disrupts the recall of old information. For example, learning a new colleague's name makes you forget the name of the person who had the desk before them.

Type	What disrupts what	Direction	Everyday example
Proactive interference	Old disrupts new	Old → new	An old password interfering with a new one
Retroactive interference	New disrupts old	New → old	A new phone number crowding out the old one

A reliable memory hook: proactive reaches forward in time (old affects new); retroactive reaches backward in time (new affects old).

Both types explain why forgetting increases when we have to store lots of similar material, rather than because time alone has passed.

Interference does not affect all memories equally. It is far stronger when the two sets of information are similar to each other, because similar material is easier to confuse.

The classic evidence is McGeoch and McDonald (1931). Participants first learned a list of words until they could recall it perfectly. They then learned a second list. The similarity of that second list to the first was varied between groups, and the researchers measured recall of the original list afterwards.

Nature of the second list learned	Recall of the original list
Synonyms (words with the same meaning)	Worst recall
Antonyms (opposite meaning)	Better
Words unrelated to the originals	Better
Nonsense syllables	Better still
Numbers (or no new list at all)	Best recall

The group that learned synonyms of the original words showed the worst recall of the original list. Because the two lists were most similar in meaning, they interfered the most.

Similarity is the key variable: the more alike two memories are, the more they interfere with each other.

A common criticism of interference research is that it is done with artificial word lists. Baddeley and Hitch (1977) tested whether interference operates outside the laboratory using a real, everyday memory task.

They asked rugby union players to recall the names of the teams they had played against so far that season. Crucially, players had missed different numbers of games through injury, so for any two players the same span of time could contain a different number of games.

• If forgetting were simply due to the passage of time, players would forget at the same rate regardless of how many games they had played.
• If forgetting were due to interference, the players who had played the most games would forget the most, because each game creates a new competing memory.

The results supported interference: the number of games played (and therefore the amount of interfering material), not the time elapsed, predicted how many teams a player forgot.

Baddeley and Hitch (1977) show interference affecting a genuine, meaningful memory task, which strengthens the real-world validity of the explanation.

The second explanation is retrieval failure due to the absence of cues, also called cue-dependent forgetting. Here the memory is available (still stored in LTM) but not accessible — we cannot get to it because the right cue is missing.

A cue is a trigger that helps us access a memory. Cues can be:

• External — features of the environment (a place, a smell, a sound).
• Internal — our physical or emotional state (being tired, drunk, anxious or drowsy).

Tulving's Encoding Specificity Principle explains when a cue works:

A cue helps recall only if it is present at both encoding (when the memory was formed) and retrieval (when we try to recall it). If the cues at recall differ from those at encoding, more forgetting occurs.

The distinction between availability and accessibility is the heart of this explanation. Interference suggests the memory is genuinely disrupted; retrieval failure suggests the memory is intact and simply needs the correct cue to be unlocked.

Retrieval failure splits into two forms, depending on whether the missing cue is external or internal.

Context-dependent forgetting occurs when external cues at recall do not match those at learning. The classic study is Godden and Baddeley (1975), who used deep-sea divers. Divers learned lists of words either on land or underwater, then recalled them either on land or underwater, giving four conditions.

Learning environment	Recall environment	Cues match?
Land	Land	Yes
Underwater	Underwater	Yes
Land	Underwater	No
Underwater	Land	No

Recall was about 40% lower in the non-matching conditions than in the matching conditions. The external context acted as a cue: when it was absent at recall, retrieval failed.

State-dependent forgetting occurs when internal cues at recall do not match those at learning. Carter and Cassaday (1998) gave participants an antihistamine (which has a mild sedative effect, creating a drowsy internal state) before learning or recall. Recall was worse when the internal state at learning did not match the state at recall — for example, learning while drowsy but recalling while alert.

Context uses external cues (where you were); state uses internal cues (how you felt). In both, a mismatch between encoding and retrieval causes forgetting.

An A-level answer must evaluate, not just describe. Use these developed points to build extended-response marks.

Interference research often lacks ecological validity. Much of the supporting evidence uses artificial word lists learned close together in time under laboratory conditions. In everyday life we rarely learn similar material in such quick succession, so the studies may overstate how often interference causes real forgetting. Interference effects appear more readily in the lab than in ordinary life, which limits how far the explanation generalises.

However, interference does have real-world support. Baddeley and Hitch's (1977) rugby study demonstrated interference in a genuine, meaningful task, and other field research points the same way. This increases the external validity of the explanation and counters the artificiality criticism above.

Retrieval failure is a robust, well-supported explanation. It is backed by evidence from both laboratory and field settings (Godden and Baddeley; Carter and Cassaday) and by everyday experience — for example, walking into a room and forgetting why, then remembering once you return to where the thought started. Support across multiple settings makes it a strong account of forgetting.

The encoding specificity principle can be criticised as difficult to test. The reasoning risks becoming circular: if a cue produces recall we conclude it must have been encoded, and if it does not we conclude it was not encoded — so the principle is hard to falsify. In addition, context effects may only matter when the context difference is large: Baddeley later noted that land versus underwater is a far more extreme difference than most everyday changes of context, so retrieval failure may explain less everyday forgetting than the divers study implies.

1. Mixing up proactive and retroactive interference

Proactive is old disrupting new (forwards in time); retroactive is new disrupting old (backwards in time). A quick discriminator: if the older memory is the problem-maker, it is proactive; if the newer memory is, it is retroactive.

2. Saying interference explains short-term memory forgetting

Interference (like retrieval failure) is offered mainly as an explanation for forgetting in long-term memory, where competing memories accumulate. Attributing everyday interference forgetting to STM misses the point of the explanation.

3. Confusing availability with accessibility

In retrieval failure the memory is available (stored) but not accessible (cannot be reached without the cue). Writing that the memory has been "lost" or "erased" describes interference-style forgetting, not retrieval failure.

4. Forgetting that cues must match at encoding AND retrieval

The encoding specificity principle requires the cue to be present at both learning and recall. A cue that appears only at recall, but was not there during encoding, does not aid memory — state this explicitly to gain the mark.

5. Name-dropping studies without using them

Naming Godden and Baddeley or McGeoch and McDonald is not enough. State what was manipulated, what was measured, and what the finding shows about the explanation. Undeveloped study references score poorly on AO1 and AO3.