How Investigators Find Inconsistencies in a Story

Investigators find inconsistencies by comparing an account against other information. They first gather as full an account as possible, usually beginning with open questions such as "tell me what happened". They then clarify it topic by topic and test it against what is already known or can be established: the person's own earlier statements, other people's accounts, the timeline, and independent evidence such as records or physical material. Research on evidence-based interviewing suggests that how and when evidence is used in an interview affects how clearly any mismatch shows. An inconsistency is not proof of lying, though. Memory is imperfect, and truthful people often change details between tellings. Professional guidance treats a discrepancy as something to explore calmly and check, not as a verdict.

Key takeaways

• Investigators compare an account with itself, with other accounts, with the timeline and with independent evidence.
• UK police guidance says accounts should always be tested against what is known or can reasonably be established.
• Getting a full account first, with open questions, gives investigators more to compare.
• Research suggests the way evidence is used in an interview affects whether inconsistencies emerge.
• Inconsistency alone does not prove deception, because truthful memories change too.

What counts as an inconsistency?

In an investigation, an inconsistency is any point where two pieces of information cannot both be true, or do not fit together without an explanation. It helps to separate four kinds:

Not every difference is meaningful. A witness who remembers a jacket as dark blue on Monday and black on Friday has changed a detail, but that change may say more about memory than honesty.

How do investigators gather an account worth comparing?

A clear comparison needs a detailed account first. The College of Policing's Authorised Professional Practice on investigative interviewing, the official source of professional practice for police in England and Wales, sets this out through the PEACE framework: Planning and preparation; Engage and explain; Account, clarification and challenge; Closure; and Evaluation.

In the account phase, the guidance describes a simple structure:

1. Initiate the account with an open-ended prompt, such as "tell me what happened".
2. Support it with active listening, including letting the person pause to search their memory without interrupting.
3. Clarify and expand it by breaking the account into manageable topics and probing each one with open and specific questions.
4. Cover gaps by examining information identified during planning that has not yet come up.

The guidance notes that open-ended questions produce answers that are less likely to have been influenced by the interviewer. That matters: if an interviewer feeds details into a question, it becomes harder to know whether a later "inconsistency" came from the person's memory or from the interview itself. The difference between question types is covered in Open-Ended Versus Leading Questions.

How are accounts tested against what is known?

The same guidance lists seven principles for investigative interviewing. Principle 3 describes an "investigative mindset": accounts should always be tested against what the interviewer already knows or what can reasonably be established. Investigators are encouraged to set objectives that will help corroborate or disprove information already known, and to fill gaps by testing information through other means where possible.

In practice, that testing usually happens outside the interview room as much as inside it. At a high level, it means:

• Building a timeline. Placing each claim in time shows where accounts overlap, conflict or leave unexplained gaps. See What Is an Investigative Timeline?
• Comparing accounts. Several people's versions of the same event can confirm or contradict each other. See How Investigators Compare Multiple Accounts
• Checking independent information. Records, documents and physical evidence do not change their story, so they give a fixed point to compare against.

The guidance also stresses fairness: investigators should not approach an interview with prejudice, and should be prepared to believe the account they are given while using judgement to assess its accuracy.

Does the way evidence is used make a difference?

Research suggests it does. Psychologists Maria Hartwig, Pär Anders Granhag and colleagues have studied what they call the Strategic Use of Evidence (SUE). The basic idea is that the timing and manner in which an interviewer uses existing evidence can affect how clearly an account's fit, or lack of fit, with that evidence becomes visible.

Two experiments with mock suspects illustrate the research:

• In a 2005 study, observers judged interviews in which evidence was revealed either early or late. Observers who watched late-disclosure interviews reached 61.7% accuracy, compared with 42.9% for early disclosure.
• In a 2006 study, police trainees who had been trained in strategically using evidence reached 85.4% accuracy, compared with 56.1% for untrained trainees. The trained interviewers elicited more inconsistencies between liars' statements and the evidence.

These are laboratory findings with staged events, where the truth was known. They show that evidence-focused interviewing can help under controlled conditions. They do not mean any method reliably identifies liars in real cases.

How should an inconsistency be challenged?

Professional guidance is clear that challenge should not become pressure. The College of Policing guidance on challenging accounts says interviewers should not use a raised voice or inflammatory language, because this can lead to a breakdown in rapport. Each false account should be treated as a separate objective, using questioning to probe and summarise.

Research on false confessions shows why this matters. A 2025 review of police-induced confessions by Saul Kassin and colleagues describes how confrontational tactics can increase the risk that innocent people confess. The broader difference between these styles is explained in Interview Versus Interrogation: What Is the Difference?

Why isn't an inconsistency proof of lying?

Because truthful people are inconsistent too. In a 2013 book chapter, Ronald Fisher, Aldert Vrij and Drew Leins examined whether testimonial inconsistency indicates memory inaccuracy and deception. They note that inconsistency is the most often cited reason for discrediting other people, but question whether that is justified. Memory research shows that recollections can change through contamination, such as misleading questions.

The reverse is also true: a perfectly consistent story is not proof of honesty. A 2018 study by Jaume Masip and colleagues summarises earlier research as showing that inconsistencies across repeated interviews do not, on their own, indicate deception.

Innocent reasons for a changing account include:

• remembering more details after a first, stressful interview;
• confusing the order of similar events;
• being influenced by a question's wording or by other people's accounts;
• stress at the time of the event or during questioning;
• embarrassment about something unrelated to the investigation.

These themes are covered in Why Truthful Accounts Can Still Be Inconsistent. The practical lesson is that an inconsistency tells investigators where to look, not what they will find.

How does Murder Book use inconsistencies?

Murder Book: The Art of Interrogation is an upcoming narrative interrogation game from Gungrounds, based in Zagreb, Croatia. Players examine unreliable stories, narrative gaps and contradictions. Development screenshots show an evidence board that links statements to each other, with lines such as "I NEVER met her!" and "I had no idea" connected to a fabricated description of a gun.

That kind of board mirrors the basic idea above: putting claims side by side to see what does not fit. But a contradiction in Murder Book is best treated the way professional guidance treats one in a real interview, as a reason to look closer rather than as proof. A changed detail could point to a lie, a mistake, fear or something the character would rather not say. Working out which explanation fits is the investigation.

Murder Book is in development for PC (Steam) and mobile, with release currently planned for the end of Q2 2027.

Sources

1. College of Policing, Authorised Professional Practice: Investigative interviewing (updated 26 October 2022)
2. Hartwig, M., Granhag, P. A., Strömwall, L. A., & Vrij, A. (2005). Detecting deception via strategic disclosure of evidence. Law and Human Behavior, 29(4), 469–484
3. Hartwig, M., Granhag, P. A., Strömwall, L. A., & Kronkvist, O. (2006). Strategic use of evidence during police interviews: When training to detect deception works. Law and Human Behavior, 30(5), 603–619
4. Fisher, R. P., Vrij, A., & Leins, D. A. (2013). Does testimonial inconsistency indicate memory inaccuracy and deception? Beliefs, empirical research, and theory. In Applied Issues in Investigative Interviewing, Eyewitness Memory, and Credibility Assessment (pp. 173–189). Springer
5. Masip, J., et al. (2018). Learning to detect deception from evasive answers and inconsistencies across repeated interviews. Frontiers in Psychology, 8, 2207
6. Kassin, S. M., et al. (2025). Police-induced confessions, 2.0: Risk factors and recommendations. Law and Human Behavior, 49(1), 7–53
7. Murder Book press sheet and development screenshots, Gungrounds