In simple terms
A friendly intro before the formal notes — no formulas yet.
Is Your Memory a Video Camera or a Storyteller?
Memory does not replay events like a recording. Each time you remember, your brain rebuilds the event from fragments and fills the gaps with expectations. The same shortcut-taking mind also biases the decisions you make. Useful, fast, and reliable enough for daily life, but far from perfect.
Think of a memory like rebuilding a model from LEGO. At the event you keep only the main bricks. Later, recalling it, you retrieve those key bricks and fill the gaps with standard bricks from your general collection, your expectations and beliefs. The finished model looks plausible, but it is not an exact replica. A leading question or a piece of misinformation is like someone handing you a wrong-coloured brick during the rebuild, and it ends up looking as though it was always there.
- 1
Memory is an active reconstruction, not passive playback.
- 2
Schemas organise information and can quietly distort recall to fit expectations.
- 3
Post-event information (leading questions, misinformation) can alter, or even implant, a memory.
- 4
The same fast, shortcut-driven mind (System 1) biases everyday judgement, e.g. anchoring and confirmation bias.
Explore the concept
Use the live diagram and synced steps — play it or tap a step card to walk through.
Full topic notes
Formal explanation with the rigour you need for the exam.
The theory of reconstructive memory
Frederic Bartlett (1932) was among the first to argue that memory is not passive or accurate. He proposed the theory of reconstructive memory: we actively rebuild memories each time we recall them. On this view memories are not faithful copies of past events but reconstructions shaped by prior knowledge, beliefs and expectations. Those mental frameworks are called schemas, and they are what we fall back on to fill the gaps in an incomplete memory.
Because schemas run automatically and outside awareness, the reconstructed memory feels just as vivid and certain as an accurate one. Confidence, in other words, is not a reliable guide to accuracy, a point with real consequences for eyewitness testimony.
Memory is an active, not a passive, process.
Schemas are mental templates used to interpret new information and to reconstruct memories.
During recall we may unconsciously change a memory so it fits a schema, through processes such as:
Assimilation — changing details to fit familiar cultural norms.
Levelling — omitting unfamiliar or 'irrelevant' details.
Sharpening — adding, reordering or exaggerating details to make the memory more coherent.
Evidence for unreliability: the misinformation effect
Building on Bartlett, Elizabeth Loftus ran a series of influential experiments from the 1970s onward. Her work demonstrated the misinformation effect: exposure to misleading post-event information can distort a memory, and in some studies can implant an entirely false one. Because leading questions are a common form of misinformation, this has profound implications for the reliability of eyewitness testimony in legal settings.
Biases in thinking and decision-making
Unreliability is not limited to memory. Everyday thinking runs largely on heuristics, mental shortcuts that deliver fast judgements with little effort. A widely used framework for understanding them is the dual-process model, popularised by Daniel Kahneman, which distinguishes two modes of thinking.
System 1 — fast, automatic, intuitive, effortless. It runs constantly and produces snap judgements.
System 2 — slow, deliberate, logical, effortful. It is engaged only when we consciously reason things through.
Most biases arise when we lean on System 1 in a situation that actually needed System 2.
Anchoring bias — the first number we meet drags later estimates toward it, even when it is arbitrary.
Confirmation bias — we seek and weight evidence that fits our existing belief and discount evidence that does not.
Evaluation: how reliable are these processes, really?
A balanced conclusion is not 'memory is useless' but 'cognitive processes are systematically distortable under identifiable conditions'. That precise claim is exactly what a top evaluation states, and what a leading question, a schema, or a System 1 shortcut can each be shown to trigger.
Strong experimental control — studies like Loftus & Palmer isolate one variable (the verb) and establish cause and effect, which is why we can attribute the change in memory to the leading question.
Low ecological validity — watching film clips in a lab is not the stress and stakes of a real crime. Real eyewitnesses may attend differently, so lab findings may overstate everyday unreliability.
Gist versus detail — memory for central, emotionally significant information (the gist) is often accurate even when peripheral details are distorted, so 'memory is unreliable' should be qualified.
Individual and situational factors — age, prior schemas, emotional arousal and confidence all moderate how much distortion occurs; unreliability is not uniform.
Real-world value — despite limitations, this research has reshaped legal practice around eyewitness testimony and police interviewing, showing the effects are robust enough to matter.
Common mistakes examiners penalise
Describing a study but never linking it — the single most common reason an otherwise good SAQ stalls in the 4–6 band. Always add 'this shows that... [claim in the question]'.
Only listing a study — naming Loftus & Palmer and giving a one-line result, with no aim or procedure, keeps you in the 1–3 band.
Getting the misinformation effect backwards — it is post-event information that distorts memory, not something that happened during the original event.
Confusing System 1 and System 2 — System 1 is the fast, automatic one; do not swap the labels or call System 2 'the biased one' (biases usually come from over-relying on System 1).
Treating schema distortion as deliberate lying — reconstruction is unconscious; the witness genuinely believes the distorted memory.
Overclaiming — writing 'memory is always unreliable'. Qualify it: peripheral details and post-event misinformation are where distortion bites; gist is often accurate.
Padding the SAQ with a second study — an SAQ needs ONE study described in depth and linked; a shallow second study wastes time the markband rewards for detail.
Worked examples
See the formulas applied — reveal one step at a time, like the exam.
Worked example 1 — Using Loftus & Palmer's (1974) first experiment, explain how leading questions can affect memory.
- 1
Aim: to investigate whether the wording of a question about a witnessed event can influence memory recall.
Worked example 2 — An investor believes a tech company, 'Innovate Inc.', is destined for success. Explain how System 1 thinking and confirmation bias could make their investment decision unreliable, using a worked scenario.
- 1
Set-up: the investor already believes Innovate Inc. is a good bet. That belief is a System 1 intuition, formed fast and held with confidence. Rather than switching on System 2 to test it, they let the intuition steer how they read the evidence, and confirmation bias does the rest.
Paper 1 SAQ: Explain, with reference to one study, how memory can be unreliable. [9]
- 1
Model answer: Memory can be unreliable because it is reconstructive: rather than replaying an exact recording, we rebuild an event at recall from schemas and can incorporate misleading post-event information. This can be demonstrated by Loftus and Palmer (1974).
How it all connects
The big idea sits in the middle — tap a linked idea to explore the link.
Tap a linked idea to see how it connects back to the main topic — that connection is what examiners reward.
Glossary
Try to recall each definition before you reveal it.
Quick check
Answer in your head first — then tap to check. No pressure.
Revision flashcards
Flip the card. Test yourself before the exam.
Reliability of cognitive processes
The extent to which cognitive processes such as memory and decision-making are consistent and accurate. Research suggests they are systematically prone to error and distortion, not simply random mistakes.
Key takeaways
Review these before you close the topic — retrieval beats re-reading.
- ✓
Memory is an active, not a passive, process.
- ✓
Schemas are mental templates used to interpret new information and to reconstruct memories.
- ✓
During recall we may unconsciously change a memory so it fits a schema, through processes such as:
- ✓
Assimilation — changing details to fit familiar cultural norms.
- ✓
Levelling — omitting unfamiliar or 'irrelevant' details.
- ✓
Sharpening — adding, reordering or exaggerating details to make the memory more coherent.
Practice — then mark it
The whole point: a real Cambridge question, marked mark-by-mark.
Get a Paper 1 SAQ marked: explain, with reference to one study, how memory can be unreliable [9]
Get a Paper 1 SAQ marked: explain, with reference to one study, how memory can be unreliable [9]
Extra simulations & links
PhET, GeoGebra and other curated tools — open in a new tab.
Frequently asked
Checkpoint
One marked question is worth ten re-reads — close the loop before you move on.
Reading it isn’t knowing it — prove it.
Before you move on: do Get a Paper 1 SAQ marked: explain, with reference to one study, how memory can be unreliable [9] on paper, snap a photo, and get examiner-style feedback on exactly where you win and lose marks.