University of Birmingham
“Eyewitness testimony is unreliable!”
“Aren’t lots of people falsely convicted on the basis of eyewitness testimony?”
“Eyewitness testimony should not be trusted!”
As memory researchers, these are comments that we encounter all of the time. Many people believe that eyewitness memory is not accurate; therefore, it should not be trusted as a basis for conviction in court. We, and other memory researchers, agree that memory can be fallible. Consequently, we agree that convicting someone on the basis of memory evidence alone is a risky strategy. Yet, we’d like to highlight that our memory may be better than many people think.
Eyewitness Testimony can be Contaminated if Collected Inappropriately
It is commonly known that memory evidence is prone to contamination if people are questioned in a suggestive manner (Loftus & Palmer, 1974). The use of leading questions to interview a witness can impact their memory evidence negatively and result in witnesses complying with the questions (Roebers & Schneider, 2000). Decades of psychological research shows that memory can be malleable when people are questioned in non-optimal ways.
Appropriate Methods of Collecting Eyewitness Memory
Despite this, there are things police officers can do to ensure that eyewitness memories are as accurate as possible. First, investigators should avoid suggestive procedures. Instead, they should use open, free recall, questions (e.g., “Tell me what happened?”) to gain accurate information, especially with children (Lamb, Orbach, Hershkowitz, Esplin, & Horowitz, 2007).
Collecting Confidence Judgements in Eyewitness Memory
Second, investigators should collect confidence judgements. A vast amount of research has emphasised that one method to understand eyewitness memory accuracy, is to collect an individual’s perception of how confident they are in the memory that they have provided (Wixted & Wells, 2017). Individuals are generally very good judges of what they remember and what they do not. Think back to the last time that you were 100% confident in your memory (e.g., yesterday I was 100% confident that I turned off the oven) and compare this to a time that you were 50% confident in your memory (e.g., today I am only 50% sure that I turned off the oven). On which occasion are you more likely to change your behaviour to verify your memory? I am more likely to go home and check the oven today, when I am only 50% confident, as there is a higher chance that I made a mistake and I did leave the oven on. Every day we automatically and effortlessly assess the strength of our memories. As a result, we are often aware of how sure we are in our own memory.
Practically, this means that the legal system can use confidence information to understand how reliable a witness’s memory evidence may be. A witness’s confidence in their memory report is generally diagnostic of the accuracy of that memory (high confidence reflects high accuracy; low confidence reflects low accuracy). This is true, even under conditions that impair memory accuracy, such as when attention is divided when forming the memory (Palmer, Brewer, Weber, & Nagesh, 2013) and when someone is intoxicated (Flowe et al., 2017). A positive relationship between confidence and accuracy still remains, because people realise that they may be less likely to be accurate in these difficult conditions, and regulate their confidence appropriately. Although many proclaim that “Eyewitness testimony should not be trusted!”, evidence suggests the opposite—with effective memory-testing procedures, eyewitness memory can be reliable.
If you wish to find out more, the Applied Memory Lab from the University of Birmingham has put together a free series of online presentations to illustrate the use of memory evidence in the legal system. The ESRC Social Sciences Festival, taking place between the 3rd-10th of November, celebrates pioneering social, economic and political research and its impact on everyday life. Our programme, ‘Memory and the Legal System’, seeks to capture the many effects of memory on the legal system, and vice versa. Presentation topics include post-traumatic stress and memory; victim alcohol intoxication and memory for sexual violence; memory accuracy in lineup parades; and innovative and cost-effective techniques for collecting and preserving memory evidence in the legal system.
More information can be found here >
You can access the presentations on our lab website starting on November 3
Join us, to learn more about the reliability of eyewitness memory.