Our minds are not like computers. Our brains do not take pictures of events; many different factors can affect the accuracy of our recall including pre-existing biases, stereotypes and emotions.
The issues of the fallibility of human memory is especially important in terms of Eyewitness Testimony as juries place such a great level of importance on the evidence presented by a witness, even if there is a possibility they may be incorrect.
Bartlett (1932) argued that memory is reconstructed from various pieces of information. This means that memories can often be inaccurate. He argued that interpretation plays a major role in the remembering of past events, and that remembering can be seen as “effort after meaning”, that is trying to make the past more logical, coherent and generally sensible.
We construct the past by trying to make it fit into our existing understanding of the world. Bartlett called this “schema” (plural “schemata”) which are packages of knowledge we store about events or places.
Any information which does not fit with our schema is simply ignored. This can have serious implications for EWT, as witnesses may choose to omit a vital piece of evidence as it does not fit with their expectations or own experiences, thus making their recall of the event less accurate and reliable.
Reconstructive Memory- Supporting
Allport and Postman (1947) investigated the effect of stereotypes on recall. A stereotype is simply a schema containing beliefs about a certain group of people. They showed participants a picture of a black person being held at knifepoint by a white man. They found that in over 50% of the cases, the final description of the picture had changed so that the black man was holding the razor. This could be explained because of people had negative stereotypes about black people and crime and this altered their memory.
Further evidence for the influence of schemas and stereotypes comes from Tuckey and Brewer who found that when participants were shown a video of a bank robbery, they recalled more details that fitted with the stereotype of a bank robber (bank robbers being male). Although supporting the theory that we are better at remembering things that fit with our stereotypes (schemata), they also remembered counter-stereotyped information. Therefore such findings contradicts the notion of schemas.
Reconstructive Memory- Evaluation
From an evaluative perspective there are some validity issues with the research discussed so far; they used laboratory experiments to test their theories which are high in lacking ecological validity as the participants didn’t witness these scenarios in real life, they looked at a picture or watched a video instead. Therefore it cannot be generalised to the wider population.
A second factor which can influence reconstructive memory (and hence the accuracy of EWT) is that of leading questions. There has been numerous research studies carried out by Loftus and her colleagues demonstrating how memory can, at times, be altered by changing the wording of a question.
One such example is that of Loftus and Zanni (1975) who found that 7% of the participants who were asked “did you see a broken headlight?” reported seeing one, whereas those asked the question “did you see the broken headlight?, 17% reported seeing one. Therefore demonstrating that leading questions can actually cause participants to remember something that wasn’t there.
However, another research study conducted by Loftus (1979) contradicts such influence of leading questions as participants were shown a series of pictures of a man stealing a red wallet from a woman’s bag and 98% were able to identify it correctly. When asked leading questions, the participants persisted in describing the purse as red, therefore it contradicts as it suggests that in some circumstances, leading questions have a limited effect on memory.
A problem with a lot of research into leading questions is that much of it is highly artificial as it takes place under lab conditions which may lead to issues of generalisability.
Yulie and Cutshall (1986) interviewed 13 people who had witnessed a real armed robbery in Canada 4 months after. They found that participants were not led by the two leading questions and that the accounts that they gave were very similar to those in their initial witness statements.
This possibly suggests that the effect of leading questions is diminished in real life situations where there are great personal consequences for the individual. Research into leading questions has led to the development of the cognitive interview which is structured to avoid leading questions and to obtain the accuracy of EWT.
Role of Emotion
A third and final factor which has been identified as influencing the accuracy of EWT is that of the role of emotion. There are opposing views on whether intense emotion hinders or helps in the formation and subsequent recall of a memory of an event. For example, Freud would claim that the effect of emotion on memory leads to a repression of the memory because it is too traumatic to recall and as a result forgets the event and details.
Role of Emotion-Supporting
MacLoed at al (1986) for example found that by investigating real life eyewitnesses’ reports of 379 physical assaults and comparing them with non-physical, there was no overall difference in accuracy between the two types of crime, suggesting that levels of emotion do not make a difference to recall.
Role of Emotion- Contradicting
Whereas in contrast, Christianson and Hubinette (1993) found that witnesses to real bank robberies who had been threatened had better recall than onlookers who were not involved due to high levels of emotion. The advantage of such evidence is that it is real life, thus meaning that it is high in ecological validity which can allow for generalisations to be made. However, uncontrollable confounding variables could have affected the results.
Flashbulb memories are vivid, long lasting memories which occur at times of heightened emotion. Brown and Kulik (1977) noticed that people were able to describe exactly what they were doing when they heard of JFK’s assassination.
Flashbulb memories were so named because it seems as if the mind has “taken a picture” of the event. They are different to normal memories as they seem to last a lifetime, are highly detailed and supposedly immune to decay.
Despite this, Talarico & Rubin (2003) found that by asking 52 students to recall their memories of 9/11 on the day after and asked about another recent event such as a birthday party or study session, consistency for the flashbulb and everyday memories didn’t differ after 1,6 or 32 weeks after. However the participants rated their recall of 9/11 as being much more vivid, and had a greater confidence in its accuracy.
The effect of emotion can be clouded if weapons are present. Research has suggested that witnesses often focus on a weapon at the expense of other details such as face or clothes. This is known as weapon focus (weapon effect).
Loftus et al (1987) found that by showing participants different slides of a customer holding a chequebook and then a gun in another, that those who saw the gun version focused on the gun and were less likely to identify the customer in an identity parade than those who had seen the chequebook version.
However, weapon focus does not always have an effect, for example, in the study by Yulie and Cutshall, the weapon effect did not happen. It is also reasonable to argue that the weapon effect may be dependent upon the length of time the crime took, for example, a quick robbery in a corner shop which lasted a minute may lead to weapon focus, whereas a prolonged hostage may give the witness longer to overcome the effects of weapon focus and thus take in more details such as facial features, tattoos etc.
It is vitally important for police and courts to understand how accurate EWT is. As the Innocence Project in America states; most wrongful convictions are due to faulty EWT.
A problem with much of research into EWT is that due to being conducted in a lab which is an artificial environment it is a world away from being a witness to a real life crime. In real life crime, emotions would be heightened and uncontrolled, unlike a laboratory experiment. Additionally, a further issue into the research of memory is that in real life, there is often no objective way of telling whether or not what the participant recalls is actually what happened.
Lastly another problem is that the majority of research into EWT are ethnocentric and thus cultural biased as they focus on Americans rather than other countries and therefore it would be difficult to apply the results found to the of other parts of the world such as Britain.