The PEACE Model, the Cognitive Interview, and Investigative Linguistics

The Reid Technique, developed in the United States by John Reid and popularised through Inbau, Reid, Buckley, and Jayne's manual Criminal Interrogation and Confessions, starts from a classification decision: is this person telling the truth or lying? Once the officer has decided a person is deceptive, the technique moves into a nine-step process designed to break down resistance and obtain an admission. Minimisation and maximisation tactics, the false-evidence ploy, and sustained psychological pressure are all part of the toolkit. The word 'interrogation' is accurate; the goal is a confession.

PEACE replaces this with a model that refuses to begin from a guilt presumption. At the Engage and Explain stage the interviewer establishes rapport and explains the purpose and format of the interview. The Account stage uses open questions to get the person's full narrative before any challenge. Challenge comes only after the account is complete, based on specific inconsistencies with known evidence, not on a prior classification of the person as deceptive. The Closure stage summarises what was said and confirms any action points. Evaluation happens after, not during.

Ron Fisher and Ed Geiselman published the Cognitive Interview in 1985, drawing on Tulving's encoding-specificity work and Bower's state-dependent memory research. The CI's original four techniques addressed a specific problem: standard interviews were retrieving far less information from cooperative witnesses than those witnesses actually held in memory, because the structure of the interview was getting in the way.

1. Mental reinstatement of context
   The witness is asked to mentally reconstruct the environment and their internal state at the time of the event: the location, the lighting, the sounds, their own feelings. This uses the encoding-specificity principle directly, using context to cue recall of events encoded in that context.
2. Report everything
   Witnesses are explicitly told to report everything they remember, including details that seem trivial or uncertain. Standard interviews signal through body language and interruption that peripheral detail is not wanted; the CI reverses this, treating all detail as potentially useful.
3. Recall in different orders
   The witness is asked to recall events in reverse order, or starting from the middle. This disrupts the tendency to produce a schema-driven narrative (what usually happens in this kind of situation) rather than a genuinely episodic one (what actually happened on this specific occasion).
4. Change perspective
   The witness is asked what another person nearby might have seen. This shifts attentional focus and can cue recall of details not accessed from the original viewpoint. It is not about speculating about other witnesses; it is a memory-access technique.

In controlled studies comparing CI to standard interview, the CI consistently produces more correct information from witnesses, typically 25-35 percent more, without a corresponding increase in incorrect details. The technique was subsequently revised by Fisher into the Enhanced Cognitive Interview (ECI), which added rapport-building, witness-compatible questioning (questions timed to the witness's own recall pace), and focused memory retrieval. The ECI is the version now standard in England and Wales and Australia.

Elizabeth Loftus began her programme of research on eyewitness memory in the early 1970s, and her 1974 paper with John Palmer, published in Journal of Verbal Learning and Verbal Behavior, remains one of the most cited in psychology. Loftus and Palmer showed participants a film of a car accident, then asked them either 'About how fast were the cars going when they smashed into each other?' or the same question with 'contacted', 'hit', 'bumped', or 'collided'. Speed estimates varied systematically with the verb. Participants who heard 'smashed' gave higher estimates and were significantly more likely a week later to report having seen broken glass in the film, even though there was none.

The mechanism is what Loftus called the misinformation effect: post-event information, including the language of questions, integrates with and alters the original memory trace. This is not about lying or poor observation: the participants genuinely remembered seeing broken glass. The implication for interviewing is direct. Every question asked of a witness before free recall is a potential source of misinformation. Closed and leading questions are especially dangerous because they introduce specific vocabulary and presuppositions that can corrupt the original memory before the witness has been asked to produce their own account.

Transcripts of PEACE/CI interviews differ from Reid-style transcripts in several linguistically measurable ways. Talk-time ratio is the most immediate: in a standard interview the interviewer often holds 50-60 percent of the speaking turns. In a well-conducted CI the witness holds 70-80 percent. This alone produces more information because information can only come from the person who holds it.

• Open-question density: PEACE/CI interviews start with broad open questions and use specific-closed questions only to clarify, not to establish new facts.
• Pause tolerance: CI-trained interviewers are explicitly trained to tolerate silence after a question, waiting for the witness to continue rather than filling gaps with new questions. Silences on the transcript are a quality indicator, not an awkwardness to be eliminated.
• Vocabulary preservation: the interviewer uses the witness's own words in follow-up questions ('You mentioned he was wearing something dark. Tell me more about that.') rather than introducing new vocabulary that might contaminate the account.
• Account completeness before challenge: in a PEACE interview the witness's full narrative is completed before any inconsistency is raised, preserving the integrity of the initial account for evidential purposes.

From a forensic linguistics standpoint, PEACE/CI interviews produce better evidence not just because they retrieve more information but because the linguistic record they create is more defensible. The absence of leading questions, the preservation of the witness's own vocabulary, and the documented sequence from open to specific to challenge all make the account more resistant to challenge on the grounds that it was constructed by the interviewer rather than recalled by the witness.

England and Wales adopted PEACE nationally in 1992 with a tiered training framework: all officers receive basic training in the model; specialist interviewers receive further training in the full ECI. Australia adopted closely equivalent guidelines, codified in the National ACHIEVE interview model in New South Wales and the VIPER model in Victoria. Canada's Guidelines for Investigative Interviewing align with PEACE principles. The Netherlands has invested heavily in CI training, partly in response to a number of high-profile false-confession cases. Norway implemented structured investigative interviewing across police forces from 2012.

The United States presents a more complex picture. Reid training remains widespread in US law enforcement, though the High-Value Detainee Interrogation Group (HIG), established in 2009, operates on evidence-based principles closer to PEACE and CI than to Reid, and publishes peer-reviewed research on interviewing effectiveness. The HIG's work has contributed significantly to the empirical literature on interview technique. Some US state police forces have adopted PEACE-style guidelines; many have not.