Learning Science · Retrieval, interference and forgetting

Accessibility and retention are not identical

Distinguish current access from retention and interpret a failed retrieval attempt without assuming complete loss.

Use the examples to connect this lesson with the earlier topics. No specialist equipment is needed.

After reading, you can

Explain accessibility
Use cues and relearning as additional evidence

A name just out of reach

A learner cannot recall a name, then recognises it immediately when it is supplied. The first attempt showed poor access with that cue at that moment. The recognition suggests some retained information, but it does not reveal exactly what was stored or guarantee future recall.

Accessibility concerns how readily information can be brought to mind under particular conditions. Retention concerns what persists from learning. Neither is directly observed as a complete internal record; we infer them from patterns of performance.

Gather another kind of evidence

Try a more informative cue, a meaningful recognition choice or a relearning task. Faster relearning may indicate retained influence of earlier experience, but it also depends on task and conditions. No single method perfectly separates a stored trace from retrieval processes.

A failed attempt may reflect weak initial encoding, interference, an inadequate cue or changes across time. Keep these possibilities open until evidence distinguishes them. Avoid telling a learner that information is definitely “in there” simply because retrieval failed; sometimes the needed content was never learned adequately.

Build more reliable access

Connect the idea with meaningful cues and practise using it in varied relevant prompts. Provide correction when retrieval is wrong or incomplete. A practical plan aims for dependable performance under intended conditions rather than proving a hidden memory exists in an abstract sense.

Support after a blank response should be chosen deliberately. A first-letter cue tests something different from a full explanation. Record which kind was required and see whether it becomes less necessary over time. The result can guide useful practice even though it cannot reveal a complete inventory of hidden memory traces.

Worked example

Returning to a formula

A learner cannot state the rectangle-area formula but explains a four-by-three array as twelve unit squares. What is retained, and what should happen next?

Inspect the explanation. The learner identifies four squares in each of three rows and obtains twelve. For one-centimetre squares, the area is twelve square centimetres. This is evidence of an accessible area relationship under a diagram cue.
Reconnect symbols to that relationship: area A equals length l multiplied by width w. With l = 4 cm and w = 3 cm, A = 12 cm². The letters label roles already explained by the array.
Later ask for the formula and an appropriate new rectangle problem, recording any diagram or hint supplied. The first blank does not show complete topic loss, while the successful diagram explanation does not prove unaided formula production.

The observed pattern supports retained conceptual knowledge with difficulty producing the symbolic form under the first prompt. It guides a targeted reconnection and later check without claiming a complete internal memory inventory.

Reconnect a quantity relationship with its symbols

A rectangle is four centimetres long and three centimetres wide, shown as three rows of four one-centimetre squares. Its area is twelve square centimetres. The relation is area equals length times width, written A = l × w. Explaining the array demonstrates conceptual performance with that cue, not necessarily unaided formula recall.

Try it yourself

Think through each question before opening the answer.

What does one blank answer establish?
Show the answer and reasoning
Failure to produce the target under that prompt and those conditions. It does not establish the complete absence of every related memory.
The response is evidence about that attempt, cue and setting. It leaves open partial knowledge, an inadequate cue, interference or weak initial learning; it cannot show that every related trace is absent.
Why avoid promising that a forgotten fact must still be stored?
Show the answer and reasoning
A failed attempt cannot establish that claim; weak encoding or genuine loss may also contribute.
Failure does not prove hidden storage any more than it proves total erasure. Complementary prompts can reveal useful retained influence, while remaining limited measures.
A learner recognises a name after failing to recall it. Can you promise they will recall it tomorrow without choices?
Show the answer and reasoning
No. Recognition under the supplied cue does not guarantee later unaided recall.
The target being visible changes access conditions. Practise the intended production and check it later with the relevant prompt.

What to carry forward

Current access and retained influence differ.
Use complementary checks cautiously.
Aim for reliable access in the intended task.

References

Sources supporting the explanations and learning checks in this lesson.