To evaluate Bartlett’s study, it provided evidence for the reconstructive nature of memory. However the study only involved seven women and 13 men. This was a weakness as it was not a representative sample of a population. The participants were tested at different time intervals; this also can be seen as a limitation. Bartlett stated that schemas only affected retrieval; however evidence suggests that schemas affect comprehension – encoding and storage. The study has low mundane realism as individuals were not asked to recall a story from a different culture in everyday life.
Moving on, stereotypes are cultural expectations; which are contained within schemas. A witness to a crime may describe what the suspect was wearing based on their criminal schema, rather than the actual clothes that the criminal was wearing. This explains why eye witness testimony’s can become distorted easily. Tuckey and Brewer (2003) asked participants what their schemas were for bank robberies. They then showed participants a video of a bank robbery and tested their memory of the event. The findings illustrated participants having a better recall for details consistent with their bank robbery schema. This suggested that schemas improve memory, or provide a framework for organizing information. When information was ambiguous, participants interpreted the information by using their schemas.
Bransford and Johnson (1972) investigated the importance of schemas. They found that if you do not have a schema, then it is difficult to understand information. A passage was read to two groups. One group was given the title (schema) and the other was not. The participants were then required to recall as many pieces of information from the text as possible. The findings confirmed that the group that did not have the schema recalled less items, whereas the group that did, recalled many more.
Further, the effect of leading questions was enquired by Loftus & Palmer (1974) and Loftus & Zanni (1975). Leading questions are directive and aimed at producing a particular response, for example, “You are willing to work every other Saturday morning aren’t you?”
When used in eye witness testimonies for court, they can alter what the witness remembers about an event and can easily distort a witnesses’ memory by changing just one word in a question.
Loftus & Palmer wanted to investigate the effects of leading questions and the accuracy of speed estimates in a car crash. The participants involved in the study were shown a film of a car crash. The participants were then asked the following questions:
How fast were the cars going when they…into each other?
The missing word was one of the following:
- Smashed
- Collided
- Bumped
- Hit
- Contacted
The effect of using different words on speed estimates exhibited that the estimated speed depended on the verb used in the critical question.
Average results:
One week later the subjects were asked if they had seen any broken glass. Although there was no broken glass in the film, 32% of subjects in the ‘smashed’ condition reported seeing glass, compared to 14% of those in the ‘hit’ condition.
To evaluate, the research lacked mundane realism; for what the observers saw in the laboratory, would not have had the same emotional impact as a real – life accident. The experiment contained low ecological validity, as the study was conducted in a laboratory, but it would be unethical to conduct this study in a real – life situation.
Participants were aware that something interesting was going to be carried out; this may have meant that they were paying full attention to it; real – life eyewitnesses are taken typically by surprise. They often fail to pay close attention to an event/ incident. Demand characteristics may have led to participants guessing the purpose of the study and behaving in ways they felt the experimenter expected, rather than behaving ‘normally’.
Participants were of a certain age – students, therefore not representing the population. A strength displayed in the research proves that memories of eyewitnesses can easily be distorted. The main distortion, relatively unimportant information (broken glass) has proved harder to produce distortions for information of central importance e.g. a weapon used by a criminal. Participants witnessed a brief film, which may have contained much less information than would be available when observing a real life incident/ crime.
Loftus & Zanni (1975) carried out a similar study to that of Loftus & Palmer. Participants were shown a film of a car crash; one group of participant’s was asked if they had seen a broken headlight, the second group were asked if they had seen the broken headlight.
17% of participants in the second group said they had seen the broken headlight, even though there was NO broken headlight in the film.
Following research into eyewitness testimony, the Home Office has issued guidelines for police to follow when interviewing witnesses/suspects. These were based on the work of Geiselman et al (1985), who developed the cognitive interview.
The cognitive interview includes, eyewitnesses being asked to mentally recreate the situation, when the crime occurred, including the environment (external state) and their mood (internal state) – context and state dependent retrieval. Being encouraged to recall everything that they can remember about the crime; recalling the crime in a different order; and lastly recalling the crime from a different perspective i.e. what do you think A could see from his/her position?
To conclude these studies I have mentioned all illustrate how fragile and unreliable eyewitness testimonies can be. The importance of eyewitness testimony in court should therefore be sincerely questioned and it is essential for police to be very careful when questioning witnesses to preserve accuracy of memory and to avert injustice.