Whilst working in America German psychologist Hugo Munsterberg was the first to suggest that psychology be applied to law, business and education. He drew attention to the capacity of perception and memory, claiming that psychological knowledge provided insight into the reliability of eye witness testimony (EWT), and in so doing, created the position as expert witness for the psychologist, Munsterberg (1908). Simultaneously this fashioned fierce opposition from the legal profession and Munsterberg was effectively silenced. In 1932 Bartlett’s ‘Theory of Reconstructive Memory’ gave further insight into the reliability of EWT by suggesting that recall is subject to personal interpretation, reliant on existing knowledge, cultural norms and the way in which we make sense of the world. He claimed that schemas (mental units of knowledge that match up to habitual contacts, objects or circumstances) can distort unfamiliar or unconscious information in order to match it with existing mental representations, therefore postulating that eye witness reports can be unreliable. It wasn’t until the 1970’s that research really got underway when psychologists began conducting experiments intent on understanding the errors more profoundly. They focused largely on problems that arise from; retrieval techniques, identification, social influence and the malleability of memory. In 1972 Loftus started the debate of the misinformation effect, demonstrating that original information is displaced or transformed. She also brought attention to misleading questions, claiming that they seal off undesirable alternatives and steer a person in a desired direction. Consequently legal organisations were advised that specific procedures used to obtain reports can either weaken or improve accuracy, as a result many organisations have adopted the use of techniques such as the Cognitive interview. More recently DNA exonerations have called fresh attention to the fallibility of eye witnesses. Many criminals previously convicted on eye witness evidence have since been found innocent by DNA analysis. Regardless of these findings EWT remains a vital part of the criminal justice system and remains the most abundant form of evidence. This paper explores how cognitive psychology has investigated and explained the complications of eye witness testimony with an enhanced focus on the contextual influence of stress and anxiety, misleading information, and the cognitive interview.
In researching the accuracy of EWT psychologists have focused their attention on three main areas: initial observation of the incident (encoding), the period between seeing and recalling (storage), and reporting or giving testimony (retrieval). Empirical research has tested many variables such as; the status of interrogator and the matter of power relations, the type of crime - violent or passive, individual differences, emotional states, age of witness and interrogational disparity. Using the principle behind the ‘Encoding Specificity Theory’ (Tulving, 1983) some techniques allow the process of retrieval to take place in a similar psychological context to that in which the information was first encoded on the belief that memory will be more accurate if cues from encoding are available at retrieval. This can be problematic since some contextual elements are difficult to reconstruct. Stress is one of them. In a stressful situation such as a violent crime the witness’s attention may narrow to the central details of the event thereby producing less reliable memory for peripheral detail. Central detail may feature a dangerous weapon, perhaps a gun or knife (weapon focus), and in doing so the witness pays attention to the weapon, excluding background features such as colour of clothing or a vehicle. This assumption is supported by evidence from a study by Loftus, Loftus, & Messo (1987). Participants listened to an argument between two people, one condition sounded more violent than the other. In the calmer condition a man appears with a pen in hand. In the violent condition a man appears with a knife saturated with blood. Participants could accurately recall the identity of the man holding the pen (49%) but recalled the knifeman’s identity considerably less (33%). In a follow up study eye movements were recorded. Loftus found that the focus of attention was the knife, diverting attention from the identity of the perpetrator. But are findings valid? The use of video clips doesn’t reiterate the emotion felt in a real life event. In a real event such as this a witness may even fear for their security and retain detail consciously or unconsciously.
Deffenbacher, Bornstein, Penrod and McGorty (2004) suggest that anxiety is influential on recall. In meta-analysis of studies Deffenbacher et al., found that heightened emotion had led to less accurate recall by witnesses. In contrast Christianson and Hubinette (1993) claim that high levels of anxiety can actually improve recall. Using a questionnaire they tested 110 victims and witnesses to real life robberies and found that those who had been victims had a more accurate recall of the event than those who were witnesses. But how can the anxiety levels of the victims compare with those who were witnesses? The field of vision, distance and overall impact from the same event were conceivably different for each individual. Moreover, could a witness be more anxious than a victim in some circumstances due to individual traits and personal experience? The term ‘Inverted U relationship’ proposes that stress and anxiety increase performance up to an optimal point, as it increases further performance drops and memory recall drops in union.
Contradicting evidence from other studies play down the significance of stress as influencing eyewitness memory. Using a real life gun shooting outside a gun shop in Canada, Yuille and Cutshall (1986) demonstrated that witnesses had amazingly accurate memories. The thief who robbed the store was shot dead. Police interviewed witnesses, then after five months a dozen of them were interviewed again. Recall was found to be accurate, even after a long period; moreover the two misleading questions inserted by the researchers bore no effect. A point to consider: The witnesses who experienced heightened stress levels where at a closer distance to the shooting, again this might have helped what they remembered due to visual advantage of the event. So why the difference in results? One possible explanation is that in an experimental setting stress is manipulated but other factors remain constant whereas in a real life situation all factors co vary.
Although evidence from research has enhanced our comprehension of how the type of crime can impact upon memory and the kind of things that are remembered, research has also established how recall can be distorted, as a consequence this provided an opportunity for psychologists to assist the legal system in an attempt to unravel these issues. It is believed that what is encoded during acquisition is critical because it forms the basis for what is stored in memory and eventually retrieved when giving testimony. Not necessarily so, Loftus and Palmer (1974) conducted a lab experiment on a sample of 45 university students. Participants were shown slides of a road traffic accident and later questioned about the speed of the car. Variations of language were used in relation to speed; smash, hit, collided, bumped, contacted. Participants who heard the word ‘smashed’ approximated a higher speed, those who heard the word ‘contacted’ approximated a low speed, suggesting that post-event information can have a significant effect on memory distortion. To test modification of memory storage and retrieval, after one week participants were asked if they had seen broken glass. The participants who had received the word ‘smashed’ as post-event information recalled seeing glass more than the participants who had received the word ‘contacted’. No glass was present. Findings suggest that the content and form of a question can lead a witness to the desired answer, thereby contradicting the findings by Yuille and Cutshall. Validity of Loftus and Palmer’s study is not without question; a lab experiment is not a good representation of an eye witness, the scenario omits contextual factors such as stress and anxiety (as discussed earlier), in addition, Loftus had a tendency to use a forced choice questionnaire format in which participants have to choose between one option or another. Recall might be more precise with an open choice or essentially opt out of giving an answer altogether.
Age factors and the time between an event and retrieval are also considered consequential to memory accuracy. It is postulated that older adults have a less accurate memory on recall than younger adults in particular when there is a longer time lapse between the event and recall (Memon, Hope, and Bull, 2003). Anastasi and Rhodes (2006) tested the possibility of age effect using three age groups; 13-29yrs, 30–40yrs and 50–60yrs. Each participant was shown 24 photographs of mixed age groups. Later they were shown 48 photographs, 24 the same as before and 24 different photos. They had to state which they had been shown earlier, conclusion: the two lower age groups were better at recall than the 50 – 60 Year olds. However, all three groups were better at recognising photographs of their own age group. A similar rule also applies to race. We do better at recognising people from our own ethnic group than from others, therefore a possible complication for giving eye witness testimony of individuals dissimilar to our own ethnic identity.
Akin to many social influences, interrogational techniques play a huge part in what gets reported accurately. It is known that steering responses and implanting fragments of information can distort memory, not to mention that memory itself is imperfect, thus the way in which people are led, probed or coerced into remembering during questioning or interrogation can have grave consequences on the party’s concerned. Standard interview procedure has been largely replaced by a technique known as the cognitive interview. The technique was first devised by Geiselman (1985). In his research participants viewed a film of a violent crime, after 48 hours they were interviewed by a policeman using one of three methods: the CI, the standard interview, and a hypnosis interview. Results verified evidence in support of the CI as witnesses recalled more relevant information during the CI method than in the other two types of interview. Four components make up the technique; context reinstatement, (return to the context in which the experience occurred), reverse order (start from the end or middle rather than beginning), change of perspective (looking at it from someone else view) and reporting everything. The interviewee is asked to recall the scene envisage who was there, what they could see and hear and what happened leading up to the event. The procedure is uninterrupted, non- leading and void of time limitation. This is thought to enhance a witness’s confidence to report details that might otherwise be deemed irrelevant or overlooked. Moreover it is assumed that when a detail is retrieved from memory it provides extra cues that might jog the witness’s memory via association.
An interviewer may then use these details to request further elaboration of specific phases. Further supporting evidence of the CI came from Fisher (1990) by applying an enhanced CI, (which makes use of more social features) to genuine police settings. Detectives in Florida were trained to use the technique with crime witnesses; results found that its use increased the amount of correctly recalled information. In the UK a survey carried on policeman confirmed prevalent use of the CI; however, while officers found it useful, they expressed concern about the amount of incorrect recall, and the amount of time it took to complete analysis of the interview. Furthermore officers declared that they were only using two of the four components, Kebbell, Milne, & Wagstaff (1999). Could this have limited the technique’s full potential? Further research might look at ways in which all features of the CI can be revised to enable quicker analysis.
The eyewitness has been the focus of this paper however, why is misrepresentation recognised as the sole fallibility of the witness, what about the people hearing the evidence? The jury and even some judges are often unaware of the potential unreliability of eyewitness accounts; might they give more weight to such testimony than necessary? Over the past century cognitive psychology has come a long way in providing explanations’ concerning the difficulties with eyewitness accounts. Advances through research have demonstrated the capacity and limitations of human memory when reporting different types of crimes within their varied context. Psychologists have shown how a mere phasing of a sentence could possibly lead to a conviction, and how our ethnic origin might influence who we recognise from a crime scene. Not only has this been of great importance to the legal system, theories and strategies have also been functional in education to assist learning, in commerce and media using surveys, and of immense interest to the wider population.
© Gisela Gina 2011
No comments:
Post a Comment