Eye-Witness Testimony Until now we have focused upon theoretical psychology that employ methods that are removed from real life

Topic 3 – Eye-Witness Testimony

Until now we have focused upon theoretical psychology that employ methods that are removed from real life. Neisser was one psychologist who criticized his fellow psychologists for concentrating too much on theoretical concepts and ignoring the practical issues involved about memory. It is memory from real life experiences that psychologists must concentrate on and one aspect of this is known as the eyewitness testimony.

Reliability of Eyewitness Testimony

It is without doubt that eyewitnesses to a crime are one of the most important people to the police when trying to get a conviction but we must remember that sometimes they can be un-reliable. Sometimes we cannot recall the exact events that surround the incident or sometimes it can be distorted so that details are lost or inaccurate ones added.

In real life it is not necessary to precisely recall specific things in order to use them in every day life e.g. we do not need to remember exactly what a 10p piece looks like in order to use it to buy a sweet. In other words there are many areas in everyday memory that we do not need to recall exactly.

Reconstructive Memory

Reconstructive memory basically means what affect the event had on us rather than the precise details surrounding it. However we must remember that reminiscing on particular events can lead to them being distorted through our prior knowledge and expectations and so care must be taken in dealing with reconstructive memory.

Bartlett (1932) carried out pioneering research in this field and it was his findings that have helped us to understand it better. He argued that we do not record memories passively i.e. like we take photographs. He believed that instead of taking exact replicas of the initial stimulus, we weave it with existing knowledge and experience to form a reconstructed memory. This is known as effort after meaning.

Bartlett carried out a number if experiments to investigate how people recall things. In one of his best-known study’s he read English participants a folk tale derived from Red Indian culture called “The war of the ghosts.” This was an unusual story for people from a western culture to understand because it contained unfamiliar supernatural concepts. After an interval, the participants were asked to recall as much about the story as possible. Bartlett found that their accounts were distorted in several ways with were consistent with a western-world view. Specifically he found the following differences:

Rationalizations – People added information or justification for actions that were not in the original.
Omissions – Information was left out particularly that of which was most difficult for westerners to understand.
Changes of Order – Events were reordered in order for the tale to make more sense.
Alterations in Importance – Certain parts were given more prominence than in the original story.
Distortion of Emotion – People incorporated their own feelings and attitudes towards the story.

So, according to Bartlett we store memories in terms of our past experience or schemas. (See below.)

Schemas

Schemas are knowledge packages which are built up through past experience of the world and which can aid us in the interpretation of new information. They are ideas that we develop about

What things are
What they are for
What we can do with them
What they mean
How they relate to other ideas

Schemas allow us not to waste attentional resources on mundane, expected features of settings and allow you to concentrate on more important things. The researcher Cohen (1993) has suggested five ways in which schemas might affect memory

This is a preview of the whole essay

Schemas

Schemas are knowledge packages which are built up through past experience of the world and which can aid us in the interpretation of new information. They are ideas that we develop about

What things are
What they are for
What we can do with them
What they mean
How they relate to other ideas

Selection and Storage – Inputs from the environment not contained within a current schema are likely to be ignored.
Abstraction – Schemas allow us to get the point if incoming material rather than precise details of it.
Interpretation and Integration – Schemas provide past knowledge to interpret current situations. This can help us with things not present by filling in the missing gap.
Normalization – Memories of events are distorted to fit in with prior expectations.
Retrieval – Schemas can aid retrieval by offering a basis for an educated guess.

This shows that schemas are useful in our every day lives as they make our experiences more predictable and use out attentional resources effectively although as Bartlett has said they can sometimes lead to distortions in memory.

Schemas and Memory Errors

Brewer and Treynes investigated the effects of visual memory by asking 30 participants, one at a time, to wait in a room for 35 seconds. The room was designed to look like an office and contained 61 items in it. Some of these items were items that you would expect to fit into an office-based schema whereas some which were incompatible with this schema i.e. pair of pliers and a brick. In a sudden unexpected recall test participants were more likely to recall items typically associated with an office-based schema such as a desk or a lamp whereas recall of the less unexpected items such as a pair of pliers and the brick was lower. Most of these errors in recall were called substitutions i.e. the participants tended to falsely recall items that were not actually there based on their own schema expectancy. These findings suggest that the participants were using schemas to ensure the rapid encoding of the visual information during their 35-second stay. At the retrieval stage, the schema influenced recall so that typical items were recalled, even if they were not actually present. Brewer and Treynes also included some bizarre items such as a skull, which was recalled surprisingly frequently. This is not so easy to explain within schema theory however a revised one put forward by Graesser and Nakamura (1982) accounts for this apparent anomaly. According to this schema-plus tag theory, we store a memory for a specific event with the general schema for that event and we attach to the schema a tag or marker to indicate any particular unexpected or remarkable aspect.

List (1986) has shown that schemas can also affect recall of events. She asked people to rate the probability of events associated with a shoplifting scenario. She then compiled a video different shoplifting acts each of which incorporated some of the events rated as high probability and some that depicted low probability. She showed this video to a completely new set of people one week later and found that they were more likely to recall high rather than low probability events. Also if they were to make inclusion errors these would be ones associated with high probability events.

Evaluation of Research on Reconstructive Memory

Bartlett has been criticized for his choice of folk tales in his studies on reconstructive memory, which he assumed would be less meaningful to people from different cultures. He did not however have an objective measure of meaningfulness. There seems little doubt that we do use stored knowledge and past experience to make sense of new information and that memories can be distorted as a result of this. However the main concerns of this theory are mentioned below

It is not clear at what stage selection, abstraction, integration and normalization occur.
The concept of a schema is too vague to be useful
It overemphasises the fallibility of memory and does not take into account that quite complex events can sometimes be remembered accurately and in precise detail.
It fails to explain how schemas are formed in the first place.
There is no explanation of how the correct schema is activated to interpret new information.

Loftus’ Research into Eyewitness Testimony

Elizabeth Loftus is one of the leading researchers in Eyewitness testimony with her being more interested in what happens to memory after a particular event rather than before it. In a typical experiment participants are shown a film or slides depicting a car crash. In the interval before recall was asked she provided them with information that either conflicted or supported the original witnessed event. For example she showed 150 participants a film of a car crash and after split the participants into two groups who were then asked two questions about what they had seen. The first group was asked questions entirely consistent of what actually had taken place. The second group was asked exactly the same questions with the exception of one – “how fast was the sports car going when it passed the barn on the country road.” This question was misleading because there was no barn in the video shown to them. The following week participants were asked a further ten questions with the final one being “did you see a barn.” Loftus found that 2.7% of the people in the first group gave the incorrect answer “yes” whereas 17.3% of the second group i.e. those given the misleading question answered “yes.” This led her to conclude that, for these people, the non-existent barn had been added to the original memory representation at the question stage so that it is now recalled as being part of the original event. It seems that people can be misled by false information even when it is discredited.

In another research method Loftus demonstrated that the precise wording of questions also has an affect in prompting false memories. Loftus and Palmer (1974) showed participants a film of a car accident and asked them “how fast were the cars going when they hit each other?” Other participants were asked the same type of question except that the word “hit” was substituted with the words “smashed, collided bumped or contacted.” They found that the word used affected the speed estimation with “smashed” producing the highest estimation and contact producing the lowest estimation. A week later participants were asked if they had seen broken glass with the participants in the smashed category giving the false answer “yes.”

Of course in real life events, which may have to be recalled later in a court of law

Happen unexpectedly and in an atmosphere of tension, which is difficult to reproduce in laboratory situations and so must be taken into account when evaluating these research methods. It has been said that fear and anxiety can also affect the way in which we remember things.

Christianson and Hubinette questioned 110 witnesses who had between them witnessed a total of 22 genuine bank robberies as opposed to set up ones. Some of the witnesses had been onlookers who just happened to be in the banks at that precise moment in time whereas others had been direct threatened in the robberies. Results showed that victims were more accurate in their recall and remembered more details than that of the onlookers. This superior recall continued to be evident even after the 15 month interval which led them to conclude that people are good at remembering highly stressful events if they occur in real life rather than artificial surroundings found in laboratory situations.

However it is also true that witnesses are not always susceptible to misleading post event information. People are generally not misled by information that is blatantly incorrect. Loftus (1979) showed participants the theft of a large red purse from a handbag. In an immediate recall test she found that 98% of the participants correctly remembered the colour of the purse. They were then asked to read an account of the theft written by a leading psychologist (this was done to lend weight to the account.) One of these accounts contained errors relating to unimportant items in the slide sequence e.g. the wrong colour was given to items not central to the action. The other account contained, in addition to these minor errors, the more glaring statement that the purse was brown. In a following recall test all but two of the participants resisted the suggestion that the purse was brown and stuck with their correct choice of red. This suggests that memory for information that is particularly noticeable and salient at the time is less susceptible to distortion. In other words people can ignore new information under certain circumstances and so maintain their original memory representation intact.

Evaluation of Loftus’ Research

In light of Loftus’ research it seems clear that memory for events can be fundamentally altered in light of post event information. However her studies have been criticized for being too artificial. It is extremely difficult to reproduce events in the laboratory that are similar to that of ones in real life and it is quite possible that eyewitnesses remember them differently as Christianson and Hubinettes study suggest.

It is also possible that participants in experiments are less accurate than people in real life events purely because they know that inaccuracies will not lead to serious consequences unlike that of real life e.g. a trial. Foster et al (1994) tested this theory on participants observing footage of a burglary. They were asked to pick out the offender in an identity parade with half being told the event was real and half that it was fictitious. Recall was found to be higher with those that were told the event was real which suggests that consequentiality was a factor.

Her research methods have also been criticized in terms of the way they are conducted. A lot of the time Loftus used what is known as “force-choice” testing. For example she would give participants two slides to choose from which can sometimes lead to false results. Korriat and Goldsmith have shown that witness accuracy can be dramatically increased if tests do not rely on force-choice testing and witnesses are allowed to give no answer if they are unsure in any way. It also seems fair to say that witnesses are able to produce far more accurate memories for events if they are given the appropriate cues i.e. recreate the context, report every detail, and recall the event in different order and changing the perspective

Loftus has also been criticized for her explanations of post event information. She believes that in the light of new information old memory traces are deleted and replaced by the new false memory. Other researchers believe that the memory trace is still available even though it has been obscured by new information.

Memory for Faces (face recognition)

We have so far considered memory in terms of events but people are sometimes called upon to give descriptions of people involved in these events. Descriptions might include things like build, hairstyle, clothes and also the context that these people were seen in i.e. you are less likely to recognize your doctor in the swimming pool. Young and Bruce (1991) call this the “little red riding hood effect.” However face recognition is viewed as un-reliable partly because of the surprise element involved when witnessing a crime and partly because the individuals are removed from the context in which they originally witnessed the crime. It seems that in most crimes the face of the perpetrator is not the focus of attention. In order to increase the accuracy in which we recall things we need greater exposure to them from different angles because our memory for familiar faces has a more 3-D quality about it.

Eyewitness Summary

Although it is possible under certain circumstances that people can recall certain events vividly and in accurate detail eyewitness accounts can be unreliable. It is therefore important for psychologists to investigate the reason why inaccuracies arise and to suggest ways they may improve.