Factors affecting the accuracy of eyewitness testimony (EWT)
Eyewitness testimony (EWT) concerns the accuracy of recall of those present at an event when it occurred. It is especially important in courts of law.
Bartlett (1932) detailed how schemas, ways of perceiving the world formed from experience, affect EWT, as memories are not accurate ’snapshots’ but reconstructions of what we believe happened in an event based on previous experience, stereotypes, mood etc.
Misleading information can affect EWT, first through leading questions, which suggest a certain answer to a witness, and also through post-event discussion, where misleading information is added to a memory after an event has been witnessed.
The witnessing of real-life events can often involve anxiety, which can severely affect the quality of recall. The Yerkes—Dodson inverted-U hypothesis explains how low and high levels of anxiety are both associated with poor recall in terms of detail and accuracy of events witnessed, while moderate anxiety is associated with good recall.
Anxiety can also affect the quality of recall through repression, where traumatic events become hidden in the unconscious mind so that a witness is unaware of them and cannot recall them. Repression, however, is a controversial idea and few psychologists see it as a valid concept.
Fig 2.5 This image of Bugs Bunny was used in the study by Loftus & Pickrell to produce a fake memory
Loftus & Pickrell (2003) investigated whether false memories could be created through the use of post-event information. 120 participants who had visited Disneyland as children were placed into 4 groups. They were asked to evaluate some advertising copy about Disneyland and answer questions about their visit there. Group 1 read fake copy featuring no cartoon characters; Group 2 read the fake copy featuring no cartoon characters but there was a large figure of Bugs Bunny (a Warner Brothers character) in their room; Group 3 read the fake copy which now featured Bugs Bunny; Group 4 read the fake copy featuring Bugs Bunny and also had the large figure in the room. 30 per cent of participants in Group 3 and 40 per cent of participants in Group 4 recalled meeting Bugs Bunny at Disneyland — some even recalled having their photo taken with him. This suggests that post-event information can be misleading so that false memories are created.
• Ginet & Verkampt (2007) found that participants made moderately anxious by being told that fake electrodes on their bodies produced electric shocks in response to incorrect answers had better recall of a traffic accident viewed on video than participants with low anxiety through being told the electrodes simply monitored bodily activity. This supports the inverted-U hypothesis.
• Loftus & Palmer (1974) found that participants’ estimates of car speeds viewed on a video were affected by which verb they were given in a question asking ’How fast were the cars going when they contacted/hit/bumped/collided/smashed each other?’ This illustrates how misleading information in the form of leading questions can affect EWT.
• Koehler et al. (2002) found that participants were less able to recall stressful words than non-stressful words, supporting the concept of repression. However, Hadley & MacKay (2006) found that stressful words were better recalled, as they are more memorable, which suggests the case for repression is not proven.
Loftus & Pickrell’s (2003) Bugs Bunny study can be regarded as superior to the more famous Loftus & Palmer (1974) study, as it uses memory of a real-life event, visiting Disneyland as a child, and therefore has higher external validity.
Research into EWT has led to changes in court procedures. The Devlin report (1976) led to convictions based on uncorroborated EWT (where there is only one independent EWT) being disallowed.
Loftus has performed many studies over many years which have produced a wealth of information that has increased our understanding of how false memories can be created.
Many studies of the effects of anxiety on EWT are laboratory based and therefore not generalisable. Real-life studies often find different results, e.g. Yuille & Cutshall (1986) found that high anxiety produced excellent recall of a real armed robbery, thereby refuting the inverted-U hypothesis.
Loftus & Palmer’s findings may be due to demand characteristics, not leading questions, as participants may have given answers they thought the researchers wanted, rather than their actual recollections.
Participants do not expect to be misled by researchers, so inaccurate recall in studies may be due to participants believing researchers’ misleading statements to be true.
One practical application of research into EWT is in advertising, where advertisers use post-event information (usually through fake nostalgic images) to try and create false positive memories of products, so that we will buy them.