You are viewing a free preview of this lesson.
Subscribe to unlock all 10 lessons in this course and every other course on LearningBro.
The Cognitive Interview (CI) is a police interviewing technique developed by Geiselman, Fisher and colleagues in the 1980s as a direct, evidence-based response to the unreliability of eyewitness testimony. Where standard police interviews of the time often relied on closed and leading questions that distorted memory, the CI applies established findings about how memory is retrieved in order to maximise the amount of accurate information recalled without increasing errors. It is one of psychology's most successful real-world applications.
Key Definition: The Cognitive Interview is a structured interviewing technique that uses psychologically-grounded retrieval strategies — context reinstatement, reporting everything, recalling in changed order and changing perspective — to improve the completeness and accuracy of eyewitness recall.
This lesson covers the AQA Paper 1 Memory content on improving the accuracy of eyewitness testimony, including the use of the cognitive interview. You are required to know the four techniques of the CI (report everything, mental reinstatement of context, change the order, change the perspective) and their theoretical bases, the enhanced cognitive interview (ECI) developed by Fisher et al. (1987), and the evidence on its effectiveness — Geiselman et al. (1985) and the Köhnken et al. (1999) meta-analysis — together with the practical issues that limit its use. The topic is the applied response to the previous lesson on factors affecting EWT, and it rests on retrieval-failure theory (encoding specificity) and schema theory from the cognitive approach. You should be able to describe the techniques and their rationale (AO1), apply them to an interviewing scenario (AO2 where a stem is given), and evaluate the technique's effectiveness and practicality (AO3).
Geiselman et al. (1985) proposed four retrieval techniques, each grounded in established memory research.
flowchart TD
CI["Cognitive Interview<br/>(Geiselman et al., 1985)"]
CI --> RE["Report Everything<br/>recall every detail,<br/>however trivial"]
CI --> RC["Reinstate Context<br/>mentally return to the scene<br/>& emotional state"]
CI --> CO["Change the Order<br/>recall in reverse /<br/>non-chronological order"]
CI --> CP["Change the Perspective<br/>recall from another<br/>person's viewpoint"]
RC -.->|based on| ESP["Encoding Specificity<br/>(Tulving & Thomson, 1973)"]
RE -.->|based on| CUES["Cue-dependent retrieval:<br/>small details trigger more"]
CO -.->|based on| SCH["Disrupts schema-driven recall"]
CP -.->|based on| SCH
Key Definition: Report everything asks the witness to include every detail they can recall, however minor or seemingly irrelevant, without editing or filtering.
Key Definition: Mental reinstatement of context asks the witness to mentally recreate both the external environment (sights, sounds, weather, layout) and their internal state (emotions, thoughts) at the time of the event.
| Technique | What the witness does | Theoretical basis |
|---|---|---|
| Report everything | Recalls all details, even trivial ones | Trivial details act as retrieval cues; reduces self-censoring |
| Reinstate context | Mentally returns to the scene and emotional state | Encoding specificity principle (Tulving and Thomson, 1973) |
| Change the order | Recalls in reverse / non-chronological order | Disrupts schema-driven recall; second retrieval attempt |
| Change the perspective | Recalls from another person's viewpoint | Disrupts expectations/schemas; accesses additional detail |
The four techniques are not arbitrary; Geiselman and Fisher built them on two well-established principles of how memory is retrieved, and pairing each technique to its principle is what marks out a thorough AO1 answer.
The encoding specificity principle (cue-dependent retrieval). Memory is improved when the cues at retrieval match those at encoding. Report everything and reinstate context both exploit this: reinstating context re-supplies the original environmental and emotional cues, while reporting every trivial detail lets one recovered fragment cue the next, so that recall "snowballs" as each detail unlocks associated ones.
The principle that a memory can be reached by more than one retrieval route. A given memory trace has multiple associations, so if one route to it is blocked or exhausted, another may succeed. Change the order and change the perspective both exploit this by forcing the witness to approach the same event from a different starting point, accessing details the first, schema-driven, chronological account missed — and simultaneously disrupting the tendency to "fill in" gaps with what usually happens rather than what was seen.
This pairing explains why the components differ in power. The two techniques grounded in cue-dependent retrieval (report everything, reinstate context) tend to be the most effective (as Milne and Bull found), because they add genuine cues; the two grounded in alternative-route retrieval (change order, change perspective) are weaker and riskier, because changing perspective in particular can tip over from retrieving into imagining what another person would have seen.
The flavour of a cognitive interview is best seen in contrast with a standard one. A standard interview might ask a closed, leading question — "Was the man tall?" — inviting a yes/no answer and risking suggestion. A cognitive interview instead opens the floor and reinstates context: "I'd like you to take yourself back to that morning. Picture where you were standing — what could you see, what could you hear, how were you feeling? Now, in your own time and your own words, tell me everything you remember, however small or unimportant it might seem — don't leave anything out, and don't worry if it doesn't seem to fit." Only later, and guided by the witness's own account (witness-compatible questioning), would the interviewer gently probe specifics, never supplying details. Crucially, the change-order prompt — "Now try telling me again, but this time starting from the very end and working backwards" — provides a second retrieval attempt that often surfaces details the chronological account skipped.
Fisher et al. (1987) developed the Enhanced Cognitive Interview, adding social and communication elements to the four cognitive techniques, recognising that the interaction between interviewer and witness strongly affects recall.
| Additional element | Detail |
|---|---|
| Rapport building | Establishing a supportive relationship before questioning, so the witness feels at ease |
| Reducing anxiety | Creating a calm, unhurried atmosphere; letting the witness take their time |
| Open-ended questions | Using questions that invite full, detailed answers rather than yes/no responses |
| Witness-compatible questioning | Adapting the order and style of questions to follow the witness's own account, not a fixed script |
| Minimising distractions & not interrupting | A quiet, private setting; letting the witness speak without interruption, which preserves their train of recall |
The ECI's insight is that even excellent cognitive techniques fail if the social dynamics are wrong: a witness who feels rushed, doubted or anxious will recall less. It thus integrates cognitive retrieval principles with effective communication.
To appreciate why the CI represents an improvement, it helps to contrast it directly with the traditional "standard" police interview it was designed to replace — an interview style that EWT research (Loftus and Palmer; Gabbert et al.) had shown could actively distort memory.
| Feature | Standard interview | Cognitive interview |
|---|---|---|
| Who leads? | Interviewer-led, following a fixed agenda | Witness-led, following the witness's own account |
| Question style | Often closed and leading ("Was he tall?") | Open-ended and non-leading ("Tell me everything…") |
| Interruptions | Frequent — the interviewer steers and probes | Minimised — the witness recalls without interruption |
| Retrieval strategies | None specifically applied | Four explicit, theory-based techniques |
| Treatment of trivial detail | Often discouraged or ignored | Actively encouraged (report everything) |
| Risk of suggestion | Higher — leading questions can implant detail | Lower by design — avoids supplying information |
The contrast highlights the CI's core logic: a standard interview risks contaminating memory through suggestion and truncating it through interruption and closed questions, whereas the CI is engineered to maximise accurate retrieval (through cues and multiple retrieval routes) while minimising distortion (through open, non-leading, witness-paced questioning). This is why the CI is not merely "a better interview" but a theoretically motivated intervention that directly targets the specific weaknesses of EWT identified in the previous topic — and why its evaluation must consider both how much information it yields and how accurate that information is.
Aim: To compare the CI with a standard police interview (and with a hypnosis interview). Procedure: Participants watched films of simulated crimes and were later interviewed using one of the methods. Findings: The CI elicited substantially more correct information (around 41 items) than the standard interview (around 29 items), without a corresponding rise in errors. Conclusion: the CI improves the quantity of accurate recall over standard interviewing.
Aim: To synthesise the accumulated evidence on the CI's effectiveness. Procedure: A meta-analysis of 53–55 studies comparing the CI (and ECI) with standard interviews. Findings: On average the CI produced about a 34% increase in correct information relative to standard interviews. However, it also tended to produce a modest increase in inaccurate details. Conclusion: the CI reliably increases the amount of correct information recalled, but at some cost to precision — a trade-off that must be weighed in legal contexts.
Milne and Bull tested the four techniques individually and in combination. They found that report everything and mental reinstatement of context, used together, produced significantly better recall than any other single technique or the standard interview, while change the order and change the perspective were less effective in isolation. Conclusion: the components are not equally valuable, which is useful guidance for officers who cannot deploy all four — and which casts particular doubt on change-perspective.
| Issue | Detail |
|---|---|
| Time-consuming | A full CI takes considerably longer than a standard interview, which police under time pressure may be unwilling or unable to allow |
| Requires extensive training | Effective use depends on specialist training that is costly and time-intensive; brief training produces poor implementation |
| Used in abbreviated form | Surveys of practice find many officers use only some techniques (often context reinstatement) and skip change order and change perspective, so the "CI" as actually delivered is rarely the full procedure |
| Not suitable for all witnesses | Very young children and witnesses with some cognitive impairments may struggle with change perspective or change order in particular |
| Quantity-versus-accuracy trade-off | The increase in correct information is accompanied by some increase in errors (Köhnken et al., 1999), which matters where accuracy is paramount |
A central strength is the substantial evidence that the CI works. Geiselman et al. (1985) found the CI elicited markedly more correct information than a standard interview without raising errors, and Köhnken et al.'s (1999) meta-analysis of over fifty studies found an average 34% increase in correct information. This matters because a meta-analysis aggregates many studies and samples, giving the conclusion far greater reliability and power than any single experiment, and because the effect is consistent across studies. The implication is that the CI's superiority over standard interviewing is well established empirically — it is not a single fragile finding but a robust pattern — which justifies its adoption by police forces and its place on the specification as a genuine improvement to EWT.
Subscribe to continue reading
Get full access to this lesson and all 10 lessons in this course.