Eye-witness testimony is an important factor taken into account by juries in deciding whether defendants are guilty or not guilty. It is important, therefore, that we have some idea of how accurate these testimonies are.
Two major issues have to be considered:
4 How successful is eye-witness identification?
4 How reliable is the testimony of eye-witnesses?
There has been a large amount of research into eye-witness identification, much of it based on identification parades (line-ups). The results of this research show that eye-witness identification is extremely poor. Shapiro and Penrod (1986) suggest that there may be a large discrepancy between an individual’s appearance during the incident and later on in the identity parade.
üActivity
Make a list of ways in which appearance can be changed.
Other factors which may affect the probability of an incorrect identification are as follows:
4 The functional size of line-up: the number of people in the line-up who match the eye-witness’s description of the culprit may make identification more difficult.
4 Whether the eye-witness knows the culprit may not be in the line-up: the eye-witness should be warned of this, otherwise the eye-witness may make a guess, thinking it has to be one of them.
4 The possibility of making relative judgements: the eye-witness chooses the line-up member who is most like the culprit ‘relative to the other members of the line-up’ (Wells, 1993, page 560).
An important aspect of eye-witness testimony is the recognition of the criminal, perhaps from ‘mug-shots’ or identity parades. Buckhout (1974) showed one of the dangers of identity parades. He arranged a purse snatching by a confederate on campus and had 52 witnesses try to select the culprit from one of two line-ups of five individuals. The criminal was in the first line-up and a similar person in the second. Ten of the witnesses were unable to pick out anyone, seven selected the culprit on the first line-up but changed their mind to the look-alike on the second line-up. Twenty eight selected another innocent person from one or other of the line-ups. Only seven of the 52 witnesses picked out the right person on the first line-up and stuck to it.
For brief glances at unfamiliar faces it is the external features such as hair and chin that most influences recognition, but hair can be easily altered in different contexts deliberately or through normal growth. Studies have shown that it is also difficult to describe a face with accuracy. We do not have the vocabulary to explain someone’s appearance in words. Deffenbacher (1988) summarised five studies and found no correlation between the accuracy of a verbal description and the accuracy of visual identification. A number of methods have been tried to overcome this. The photofit technique (composite face constructed from a collection of eyes, noses, mouths, etc.) has led to poor recognition. Christie & Ellis (1981) found that targets were recognised better using verbal descriptions than from photofits. Buckout & Regan (1988) indicate that it is more difficult to recognise a person from another race than your own race.
Eye-witness identification is also complicated by the fact that we are good at knowing that we have seen a face before but not good at knowing where we’ve seen it. Brown (1977) showed participants photos of faces and a choice of two rooms they had been shown in. Recognition of faces was good but identifying the correct room was poor.
&
!SAQ 3
Why might laboratory studies exaggerate the accuracy of face recognition?
Memory is an active process: as well as organising incoming material, we also impose meaning upon it. One way of achieving this is through the use of language. Loftus & Loftus (1975) showed a film of a traffic accident and asked questions about what was seen. A week later, the participants were asked about the film again. Loftus found that the way in which the questions were asked had an effect on what the participants remembered. One group was asked immediately after seeing the film, ‘How fast were the cars going when they smashed into each other?’ The other group was asked ‘How fast were the cars going when the hit each other?’ Those participants who had heard the word ‘smashed’ replied that they remembered seeing glass scattered around after the accident (there hadn’t been any glass at all). Their memory of the accident was far worse than the other group’s memories, even though they had seen exactly the same film.
The Loftus study shows that memory
is not simply a ‘tape-recording’ of past events.
&
In a further investigation of how much eye-witness testimony is open to distortion, Loftus & Zanni (1975) showed a film of a car accident. When participants were later interviewed about the accident, they found that the wording of the questions could have a significant impact on ‘recall’. Half of the participants were asked: ‘Did you see a broken headlight?’ The other half were asked: ‘Did you see the broken headlight?’ 7% of those who received the ‘a’ question replied yes, whilst 15% of the ‘the’ group said yes. This shows how subtle changes of wording may distort recall.
Criticisms of this type of research, however, show that it lacks ecological validity as it is laboratory based and does not have the emotional impact of witnessing an incident. Also, research has focused on the recall of ‘peripheral’ details and, as Fruzetti et al (1992) point out, it is more difficult to distort witnesses’ memory for key details such as the murder weapon involved in the incident.
&
!SAQ 4
a) How does Loftus explain the effect of leading questions on EWT?
b) Under what circumstances is misleading information ineffective?
üActivity
Discuss whether being emotionally aroused would affect your memory. Would it make you remember things more or less vividly?
In fact, experimental evidence shows us that emotional arousal leads to poor recall for details. Loftus et al (1978) showed a film of a hold-up and then tested memory for details. The results showed that a high-arousal version of a young boy being shot and falling to the floor clutching his bleeding face, led to poorer recall than a low-arousal version. However, research into so-called ‘flash-bulb memories’ does show that in certain circumstances memory can be intensified by emotionally charged events.
As a result of evidence from eye-witness studies, the Devlin Report (1976) recommended that judges should instruct juries that it is not safe to convict on a single eye-witness alone, except in exceptional circumstances such as the witness is a friend or relative, or when there is substantial corroborative evidence.
Memory for people and social events is considered to be of two types:
4 Episodic: specific, concrete events, e.g. a particular bank robbery
4 Semantic: abstract, concerned with general properties and meanings, e.g. everything you know about bank robberies in general
According to Fiske and Cox (1979), when asked to remember what they can about someone, people usually start with an account of what they look like, followed by what happened and finally what kind of person he/she is. Appearance comes from episodic memory and traits from semantic memory.
The context is also important. Attempts by police to reconstruct crimes are often effective; empathising is also useful for enhancing person memory. Harvey et al (1980) found that people remember more about a person when they are asked to imagine themselves in the ‘other person’s shoes’ and try to share the other person’s feelings. (This might be useful when revising for a psychology examination. Imagine yourself taking part in the empirical investigation and this will help you to remember it.)
Memory for people may be different in significant aspects
from memory for events. Using a social-psychological approach,
We tend to recall a large amount of detail, including personality traits, attitudes and beliefs, appearance etc. We also tend to group together information which is positive and separate it from that which is negative.
It seems that the race or culture of the person attempting to recall information and that of the person about whom the information is to be recalled, can have an impact upon the accuracy of recall.
Goldstein & Chance (1985) suggest that recognising faces is a complex skill that we develop and improve upon. However, to do this we have to have experience of the faces we want to recognise. Western people tend to experience difficulties in recognising faces of Japanese people. Similarly, to many Asian or black people, whites all look very similar. But these effects tend to come from lack of experience in meeting people from the different groups. With experience, we can soon learn to be more sensitive to the differences in people’s faces.
&
!SAQ 5
The police evidence in a particular trial is largely based on identification by eye-witnesses. Explain how a psychologist might help the defence case. Justify your answer with reference to empirical studies.
Hogg & Vaughan (1996) offer the following factors for improving the accuracy of eye-witness testimony:
4 the witness goes back over the scene or the crime to reinstate additional cues
4 the witness has already associated the person’s face with other symbolic information
4 the witness was exposed to the person’s face for a long time
4 the witness gave testimony a very short time after the crime
4 the witness habitually attends to his/her surroundings
4 the witness generally forms vivid mental images
4 the person’s face was not altered by disguise
4 the person looked dishonest
It is a commonly held view that hypnosis will aid eye-witness memory. However, many studies have found no advantage from using this (e.g., Smith, 1983). Indeed, Orne et al (1984) have shown that hypnosis can actually distort recall.
Some of the reasons why evidence gathered under hypnosis should be treated with caution are as follows:
4 People may pick up on suggestions communicated by the hypnotist and incorporate these into their own memory — in effect ‘leading questions’ are more likely to produce distorted memories
4 Hypnotised people sometimes ‘see’ things that were not there and fail to report things that were there
4 Confidence with which people give information is high even though it may be incorrect. This may lead to false trails in the investigation.
4 If hypnosis makes mental images more vivid, hypnotised people may confuse these images with actual memories
[Based on findings by Laurence & Perry (1983), Rathus (1987) and Hassett & White (1989).]
The conclusion of a panel appointed by the American Medical Association was that hypnosis sometimes produces additional details that are unreliable. Police are recommended to limit its use to the investigative stage of an enquiry where it may produce clues whose details could be checked by other sources rather than accepting recall under hypnosis as evidence itself.
The cognitive interview has proved to be a better tool than the standard police interview with no more errors (Fisher & Geiselman, 1988). The technique uses 4 strategies designed to maximise recall:
4 Mentally reinstating the environmental and personal context that existed at the time of the crime.
4 Reporting everything regardless of its perceived importance
4 Recounting events in a variety of orders
4 Reporting the events from a variety of perspectives
&
!SAQ 6
Outline the advantages of the cognitive interview in improving the accuracy of EWT.