Gabbert et al took 120 participants and put them into pairs. He then showed each participant a video of a girl putting a book back into a University office. However, each person from the pairs could see an aspect of the video that the other partner could not, for example, one person could see the title of the book that the girl was returning whilst the other could see the girl stealing a £10 from a wallet. After watching the video, the participants discussed what they saw in the video with each other before completing a questionnaire independently. They found that 71% of participants stated they had seen the girl do something which they actually had not seen but just heard from their partner. This figure was 0% for the control group in which there was no post event discussion.They also found that 60% of the participants said that the girl was guilty of stealing the £10 note even though they had not actually seen her carry out the act themselves. They concluded that participants may go along with other people due to memory conformity where a person may go along with the answer of someone else because they think they are wrong and the other person might be right or they want social approval of the other person. This shows that post event discussion can have a negative effect on eye witness testimony as it can lead to people making judgements that are not entirely true.