Fake video dramatically alters eyewitness accounts
Researchers at the University of Warwick have found that fake video evidence can dramatically alter people's perceptions of events, even convincing them to testify as an eyewitness to an event that never happened.
Associate Professor Dr Kimberley Wade from the Department of Psychology led an experiment to see whether exposure to fabricated footage of an event could induce individuals to accuse another person of doing something they never did.
In the study, published in Applied Cognitive Psychology, Dr Wade found that almost 50% of people shown fake footage of an event they witnessed first hand were prepared to believe the video version rather than what they actually saw.
Dr Wade’s research team filmed 60 subjects as they took part in a computerised gambling task. The subjects were unknowingly seated next to a member of the research team as they both separately answered a series of multiple-choice general knowledge questions.
All subjects were given a pile of fake money to gamble with and they shared a pile of money that represented the bank. Their task was to earn as much money as possible by typing in an amount of money to gamble on the chances of them answering each question correctly. They were told the person who made the highest profit would win a prize.
When they answered each question, subjects saw either a green tick on their computer monitor to show their answer was correct, or a red cross to show it was incorrect. If the answer was wrong, they would be told to return the money to the bank.
After the session, the video footage was doctored to make it look as if the member of the research team sat next to the subject was cheating by not putting money back into the bank.
One third of the subjects were told that the person sat next to them was suspected of cheating. Another third were told the person had been caught on camera cheating, and the remaining group were actually shown the fake video footage. All subjects were then asked to sign a statement only if they had seen the cheating take place.
Nearly 40% of the participants who had seen the doctored video complied. Another 10% of the group signed when asked a second time by the researchers. Only 10% of those who were told the incident had been caught on film but were not shown the video agreed to sign, and about 5% of the control group who were just told about the cheating signed the statement.
Dr Wade said: “Over the previous decade we have seen rapid advances in digital-manipulation technology. As a result, almost anyone can create convincing, yet fake, images or video footage. Our research shows that if fake footage is extremely compelling, it can induce people to testify about something they never witnessed.”
Notes to editors
For more information, please contact Kelly Parkes-Harrison, Communications Officer, University of Warwick02476 150483/02476 574255, 07824 540863, K.E.Parkes@warwick.ac.uk