Menu Close
Memory preserving serum? Jorge Royan/Wikimedia Commons, CC BY-NC-SA

Our study suggests alcohol may actually protect eyewitness memory – here’s how

Ever witnessed a punch up during a boozy night out? Did you assume that you probably wouldn’t be a reliable witness because you’d been drinking? You may have been right, but our latest research indicates that in some circumstances this is not the case.

There is a strong link between alcohol and crime. According to the Crime Survey for England and Wales, 70% of public violent incidents in 2013-14 were alcohol related and 93% of those happened in pubs, bars and clubs where alcohol is sold. Given these numbers, it is likely that many witnesses and victims will be under the influence when they witness a crime. But are drunk witnesses always less reliable than sober ones?

In certain situations alcohol can have a beneficial effect on memory. Our study, published in Psychopharmacology, shows that if alcohol is consumed after witnessing a crime it can protect memory from misleading information.

The 83 participants in our study watched a video of a staged theft, where a man and a woman entered a house and stole some jewellery, money and a laptop. The robbers then swiftly left the property, before the homeowner could stop them.

After watching the film, participants were split into three groups. Members of the first group were given alcohol and were aware that they had been given it. The second group were told they would be drinking non-alcoholic beer, but in fact did drink alcohol (the purpose of this was to ensure as far as possible that it was the effect of the alcohol itself and not expectations about the effect of alcohol that would cause any effects). And the third group did not get any alcohol and knew they were not drinking.

On average, participants did not exceed the English drink-drive limit of 80mg of alcohol per 100ml of blood. All participants were then exposed to some fake information about the crime video. For example, it was suggested that the victim’s jumper was green instead of blue and the thief’s hair was brown instead of black. The following day, all participants returned to the lab when sober and their memory about the crime was tested.

Drunks are better than their reputation

While previous research suggests that alcohol may harm memory and make people more suggestible, our study found that those who did not drink alcohol were more likely to remember false information compared with those who had, whether they knew they were consuming alcohol or not.

We think this is because alcohol blocks new incoming information, including misinformation, so it is less likely to have a negative impact on what was witnessed.

Worryingly, not only were our sober witnesses more suggestible to misinformation than our alcohol consumers, but, when they were asked, they also expressed greater willingness to testify these incorrect responses in a court of law.

Our research challenges the intuitive view that alcohol is bad for eyewitness memory recall by showing that, in fact, it can be the timing of alcohol consumption that is important when it comes to determining how accurate and reliable inebriated witnesses are. Although we still don’t know the effects of different volumes of alcohol on the reliability of eyewitness testimony.

It may be too soon to tear up the rule book on inebriated witnesses. Andrey_Popov/

It is perhaps too soon for police officers, lawyers and judges to tear up the rule book regarding inebriated witnesses. We are fully aware that further research is needed to explore the exact reasons why in certain situations alcohol may protect our memory from misinformation and whether this happens in more realistic situations too, for example, when people have consumed larger volumes of alcohol than they did in our study.

The theory underpinning our findings could also explain that horrible feeling heavy drinkers get the morning after the night before, that they may have said or done something they can’t remember, as alcohol seems to prevent new information from entering our memory.

Want to write?

Write an article and join a growing community of more than 175,100 academics and researchers from 4,818 institutions.

Register now