Misleading Information and False Memories

Misleading information a witness is exposed to can, through interference, affect the witness's memory and in some cases completely destroy the value of the testimony. Examples of misleading information include newspaper notices or images the witness has seen in the media, or conversations between witnesses.

A witness's memory can also be affected by leading questions. Because of this, one must be careful about how questions are phrased during an interview. A leading question nudges the witness in the right direction, but it is not really possible to distinguish between leading and misleading questions, where misleading questions instead nudge the witness in the wrong direction. The question itself can create interference in the witness and create a false memory.

Examples of different types of leading questions:

• Wasn't there a small tractor out on the road that evening?
• Did you see the blue bicycle that evening? (compare with the experience of being asked "Did you see a blue bicycle that evening?")
• What speed would you say the vehicle had when it crashed into the house? (compare with the experience of being asked "What speed would you say the vehicle had when it came into contact with the house?")

False memories: The questions can change the answers

An experiment showing how questions can affect a witness's memory is the following: participants watched a film of a car accident and were then asked questions. Half were asked, "How fast was the car going on the country road?" Half were asked, "How fast was the car going when it passed the barn?" But no barn was ever shown in the film. A week later, 17% of those asked about the barn reported having seen one, compared with only 3% in the other group.

In another experiment, researchers staged a robbery at a train station in which the victim claimed to have had a tape recorder taken. Witnesses to the event (who did not know it was staged) were then asked what they had seen, including being asked to describe the tape recorder. Over half of the witnesses gave a description of the tape recorder even though they could never have seen it, since the "thief" only pretended to take something from the bag and hide it inside his jacket.

A third experiment demonstrating the importance of non-leading questions involved showing participants a filmed car accident and then changing only one word in a question asked afterward. "How fast were the cars going when they collided with each other" was reworded so that the word collided was replaced with "crashed into each other", "bumped into each other", "contacted each other", and so on. Comparing how fast the participants estimated the cars were going, the estimates for the milder words differed from those for the harsher words by roughly 15 km/h.

What affects the emergence of false memories?

If new information is mixed with the original memory immediately after the event, we remember correctly in about 50% of cases. If the information is instead introduced a week later, right before recall, we remember correctly in about 20% of cases. The timing of when the information is introduced is therefore decisive.

Another important factor is whether we notice that the new information does not match our own experience. If we do notice this, we can dismiss it as incorrect and retain our original memory.

This can explain why peripheral information is easier to influence than central information. It also explains why information from credible sources has a greater effect: we then tend to question our own experience rather than the source.