Case File 241 · Investigation Psychology

Why Do People Revise Their Story During Questioning?

By Murder Book Editorial TeamPublished by Gungrounds, the studio making Murder BookPublished 3 October 2026Updated 3 October 2026

People revise their story during questioning for many ordinary reasons, and most of them have nothing to do with lying. Human memory is reconstructive, not a recording, so an account is rebuilt each time it is recalled and can come out differently. Later tellings often add new details that were there all along but not mentioned at first, a well-documented pattern called reminiscence. Questions themselves can reshape what a person remembers. People also correct genuine mistakes, clarify vague wording, or recover a memory once a cue jogs it. A changed or expanded account can be a warning sign in some situations, but by itself it is not proof that someone is being dishonest.

Key takeaways

Memory is reconstructive: each recall partly rebuilds the event, so accounts can shift honestly.
New details appearing in a later telling (reminiscence) are common and are often nearly as accurate as details reported from the start.
The way questions are asked can change what someone remembers, through leading wording and post-event information.
Consistency is a weak guide to accuracy; outright contradictions matter more than added detail.
A revised story can raise questions, but it is not, on its own, evidence of lying.

Is memory a recording or a reconstruction?

Decades of research show that memory does not work like a video file. It is reconstructive: when we remember, we rebuild the event from fragments, and that rebuilding is open to influence. Loftus and Palmer demonstrated this in 1974, when they showed people films of car accidents and found that changing a single word in a later question, from "hit" to "smashed", raised their speed estimates and even made them more likely to recall broken glass that was never there.

If a single word can reshape a memory in a controlled study, it is no surprise that a long, stressful interview produces accounts that move around. Revision is partly built into how remembering works. For a fuller treatment, see how reliable eyewitness testimony is.

Why do new details appear later?

One of the most common forms of revision is the arrival of fresh detail in a second or third telling. Psychologists call the recall of new information across repeated attempts reminiscence, and the overall increase in how much is remembered hypermnesia. Both are normal features of memory, not signs of invention.

A 2013 study by Odinot, Memon, La Rooy and Millen, published in PLoS One, found that participants interviewed twice recalled a substantial proportion of new details in the second interview, and that overall accuracy stayed high even as the total amount of information grew. Repeated interviewing tended to add correct information rather than corrupt the account, especially when the interviews were conducted well.

So when a witness or suspect says something in a later session that they did not mention earlier, the ordinary explanation is that recall is still unfolding. This is also why investigators value careful re-interviewing, discussed in whether repeated recall can improve memory.

Does inconsistency mean the account is inaccurate?

Not in the way many people assume. In courtrooms, an inconsistency is often treated as proof that a witness is unreliable or dishonest. The research draws a finer line.

Stanley and Benjamin's 2016 study in Cognitive Research: Principles and Implications examined exactly this. They found that details recalled only later (reminisced details) were less accurate than details reported consistently across tellings, but just as accurate as details that were reported initially and then dropped. Their conclusion was that it is the inconsistency of recall, rather than the mere fact of adding something new, that signals lower accuracy. In other words, new detail is not automatically suspect; a genuinely contradictory account is a stronger concern than an expanded one.

This matters because consistency and accuracy are not the same thing. A perfectly consistent account can be consistently wrong, and a story that grows across interviews can still be largely correct. Judging truthfulness by whether the wording stays identical is therefore unreliable, a point that connects to why changing a story is not a dependable sign of lying.

Can questioning itself change the story?

Yes, and this is one of the most important reasons accounts are revised. Information introduced after an event, including the wording of questions, the suggestions of an interviewer or things heard from other people, can be absorbed into memory and change what a person later reports. This is the misinformation effect, and it means an interview is not a neutral readout of memory but an event that can alter it.

Leading questions, repeated pressure to add detail, and exposure to other accounts can all push a story in a particular direction without the person realising. The result can be a sincerely revised account that has drifted from the original event. These mechanisms are set out in what the misinformation effect is and how post-event information can change an old memory. Good interviewing tries to minimise them by using open questions and avoiding suggestion.

When is a revised story actually a concern?

Revision becomes more significant when it has particular features, though none is decisive on its own:

Direct contradiction of a specific, checkable fact, rather than the addition of new detail.
Changes that track the evidence. A version that shifts only after the person learns what investigators already know is worth examining.
Revisions that remove verifiable specifics and replace them with vaguer claims.

Even then, the research warns against leaping to conclusions. Stress, fatigue, trauma and simple memory error can produce all of these patterns in honest people. The sound approach is to test each version against independent evidence, not to treat any change as a confession. A behaviour, including revising a story, is a reason to ask more questions, never proof of guilt.

Is there a game about shifting accounts?

Murder Book: The Art of Interrogation, an upcoming narrative interrogation game from Gungrounds, is built around accounts that move. Rather than rewarding you for catching any change and declaring a suspect guilty, it asks you to examine unreliable stories, weigh narrative gaps and decide which revisions matter. Its press kit names Narrative Gaps and Fabrication among the behaviours players study, and development screenshots show an evidence board that links statements to one another so you can see where a story genuinely contradicts itself rather than merely grows.

That design reflects the research here: revision is often honest, and the task is to tell an ordinary added detail from a meaningful contradiction. Cases are built for focused 5–10-minute sessions. Murder Book is in development for PC (Steam) and mobile, with release currently planned for the end of Q2 2027.