Case File 260 · Investigation Psychology

Can New Information Change an Old Memory?

Yes. Information a person encounters *after* an event, called post-event information, can change what they later remember about it. When someone reads a news report, hears another witness, or is asked

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

Yes. Information a person encounters after an event, called post-event information, can change what they later remember about it. When someone reads a news report, hears another witness, or is asked a leading question, details from that new information can blend into the original memory. People then recall the altered version with sincere confidence, usually unaware that anything has shifted. This is one of the most heavily studied findings in memory research, and it is why investigators try to take an account before a witness is exposed to other sources. A changed memory is not the same as a lie: the person genuinely remembers it the new way. The effect is strongest when the new information is plausible, comes from a trusted source, and arrives before the memory has been firmly recorded.

Key takeaways

Post-event information is anything a person learns after an event that can reshape their memory of it.
In classic studies, the wording of a single question changed what witnesses later reported seeing.
The altered memory feels real; people are usually unaware their account has changed.
A changed memory is not evidence of lying, because the person believes the new version.
Investigators reduce the risk by recording accounts early and keeping witnesses apart.

How can something you learn later change a memory?

Memory is not a recording that plays back unchanged. Psychologists describe it as reconstructive: each time you recall an event, you rebuild it from fragments, and gaps can be filled with information from other sources. If some of that information arrived after the event, it can become part of the remembered scene.

Elizabeth Loftus, whose work defined this field, summarised decades of findings in a 2005 review. She describes the misinformation effect as the impairment in memory for the past that follows exposure to misleading information. The new detail does not have to be dramatic. A changed noun or a false assumption buried in a question can be enough.

The important point for an investigation is that the person is not choosing to change their story. The shift happens below awareness, which is covered in more depth in What Is the Misinformation Effect?.

What did the classic experiments show?

Two early studies are still the clearest demonstrations.

In Loftus and Palmer's 1974 experiment, participants watched film of a car accident and were then asked how fast the cars were going when they "smashed", "collided", "bumped", "hit" or "contacted" each other. The single verb changed their speed estimates. Participants given the word "smashed" gave the highest estimates. In a later session, those same participants were more likely to report having seen broken glass, although there was no broken glass in the film.

In Loftus, Miller and Burns's 1978 study, participants saw slides of an accident. A question that referred to a "stop sign" or a "yield sign" could later change which sign participants remembered seeing, even when the question described the sign incorrectly. The verbal information in the question had become integrated with the visual memory of the scene.

Neither study involved lying witnesses. Ordinary people, trying to be accurate, absorbed a detail from the questioning and reported it as something they had seen.

Does the original memory get erased?

This is still debated. One view is that misinformation overwrites or blocks access to the original memory. Another is that both versions survive and the person simply struggles to tell which detail came from where, a problem known as a source-monitoring error. Loftus's 2005 review notes that researchers have explored what happens to the original trace after misinformation appears to have made it inaccessible, and the question has not been fully settled.

What is clearer is when people are most vulnerable. Misleading information tends to have more effect when it is plausible, when it comes from a source the person trusts, and when time has passed since the event so the original memory has faded. Warning people in advance that they may have been exposed to misinformation can reduce the effect, but only under limited conditions.

Why does this matter in a real investigation?

Eyewitness accounts carry weight, and a contaminated account can point an investigation in the wrong direction. Analysis by the Innocence Project found that eyewitness misidentification was a contributing factor in a large share of DNA exoneration cases in the United States, the single most common factor among them. Memory that has absorbed post-event information is part of why confident witnesses can still be mistaken, explored in Why Can a Confident Witness Still Be Wrong?.

Good practice is built around protecting the memory before it changes:

Take a full account as soon as possible, before the witness reads coverage or talks to others.
Use open questions that invite free recall rather than leading questions that suggest an answer.
Keep witnesses apart, because one witness's account can reshape another's, as set out in Why Should Witnesses Be Interviewed Separately?.
Record the interview, so later listeners can see exactly how a detail entered the account.

Does a changed memory mean someone is lying?

No. This is the distinction that matters most. Lying is a deliberate attempt to mislead. A memory altered by post-event information is something the person sincerely believes. They are reporting their memory honestly; the memory itself is wrong.

That is why a witness who changes a detail, or becomes more certain over time, is not automatically unreliable or deceptive. The change may simply reflect what they have been exposed to since. It is also why investigators look for independent, physical corroboration rather than treating a vivid, confident account as proof on its own.

Can you tell when your own memory has been changed?

Usually not, and this is what makes post-event information so powerful. When a suggested detail is absorbed, it does not arrive flagged as borrowed. It feels like an ordinary part of the memory, with the same sense of having been there. In the classic studies, participants were not aware that a question's wording had shaped their later report; they simply reported what they now remembered.

That inability to feel the change from the inside has two consequences. First, a witness cannot reliably protect their own account just by trying hard to be accurate, because the error is not a lapse of effort. Second, certainty is a poor guide: a detail that entered after the event can feel as solid as one that was truly observed. For that reason, investigators and courts lean on external checks rather than the witness's own sense of sureness, a theme in How Reliable Is Eyewitness Testimony?.

Is there an upcoming game that explores unreliable memory?

Murder Book: The Art of Interrogation is built around accounts that do not quite hold together. It is an upcoming narrative interrogation game from Gungrounds, an independent studio in Zagreb, Croatia, inspired by real police footage and psychological interviews. Players work through unreliable stories, narrative gaps and contradictions rather than treating any single statement as settled truth.

Post-event information is one reason those gaps appear. A witness or suspect may have absorbed a detail from the news, from another person, or from how a question was framed, and now recalls it as something they saw. The game's design leans into that uncertainty: a shifting account is a prompt to look closer and seek corroboration, not an instant verdict. 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.