Eye Witness Testimony
Misleading Information:
Leading Questions
Misleading information research
Leading questions
A leading question is when the wording of the question may lead (or mislead) you to give a particular answer. This is an issue for eyewitness testimony because police questions may sometimes unintentionally direct witness to give a certain answer.
​
Elizabeth Loftus and John Palmer in 1974 conducted a study to look at leading questions.
Procedure
They had a sample of 45 participants who were divided into five groups of nine. All 45 participants watch the same film clips of car accidents and then were asked questions about the accident. The critical question (the leading question) was how fast were the cars going when they ‘hit’ each other. This question was the critical question because each group was given a different verb. One group heard the word ‘hit’, the second group ‘contacted’, a third ‘bumped’, the fourth group ‘collided’, and the final group ‘smashed’.
Findings and Conclusions
The main estimated speed was calculated for each of the five groups, and it was found that the verb ‘contacted’ resulted in the lowest mean estimated speed which was 31.8 mph. The verb which produced the highest mean estimate was the verb ‘smashed’, with a mean estimate of 40.5 mph. For the other groups, ‘hit’ produced a 34 mph estimate, ‘bumped’ 38.1 mph estimate and ‘collided’ 39.3 mph estimate.
Loftus and Palmer concluded that the verb in the question had a significant impact on the eyewitnesses recall of the event. They suggested that the wording of the question did not affect the participants memories but made the participants respond in a particular way. This is referred to as the response-bias explanation.
​
Loftus and Palmer (1974) then went on to do a second experiment to discover if leading questions can actually change a person's memory of an event, not just make them respond in a particular way. This effect is referred to as the substitution explanation, which proposes that the wording of a question can change the participant's memory of an event.
In the second experiment Loftus and Palmer showed participants clips of car crashes. In this experiment there were two conditions one who heard the word ‘smashed’ another condition where participants heard the word ‘hit’. As predicted, those who heard the word smashed when asked whether they had seen any broken glass were more likely to say that they had compared to those who had heard the word hit. This shows that a verb can alter a memory of an event, hence memory substitution.
​
​
Misleading Information: Post Event Discussion
Misleading Information Research
Post event discussion
One of the issues with eyewitness testimony being unreliable is that eyewitnesses to an event tend to discuss their experiences with each other following an event. This is referred to as post event discussion.
​
Gabbert et al in 2003 conducted a study whereby participants saw a video of a crime (someone having the purse stolen). Each participant watched a video of the crime, but each video was filmed from a slightly different angle. This meant that each participant could only see one element of the crime. After they had seen the video Gabbert put the participants into twos and asked them to discuss what they had seen. She then asked them individually to recall the event.
Findings and Conclusions
Gabbert found that 71% of the participants mistakenly recalled aspects of the event that they couldn't possibly have seen in the video that they had watched.
She concluded that this error in memory was because they had had the opportunity to discuss what they had seen with another person and that the memory had therefore become contaminated with the memory of the person who they discussed the event with.
She had a control condition where there was no discussion at all. In this condition there was 0% inaccuracy. She concluded that these results were because of memory conformity. This is where people will go along with each other often in order to feel socially approval. They believe that the other witness is right and they are wrong.
Memory conformity means that the actual memory is unchanged.
Memory contamination, however, is when the eyewitness testimony may become altered or distorted when the witness combines information from other witnesses with their own memories. This means that the memory is changed.
​
Evaluation of Research into Misleading Information
1. One strength of research into misleading information is that it has important practical uses within the criminal justice system. Inaccurate eyewitness testimony can have a serious consequences. This has led to the police being very careful about how they phrase questions when they are interviewing eyewitnesses. This shows that psychologists can help to improve the way the criminal justice system works by protecting people from faulty convictions based on unreliable eyewitness testimony.
​
2. One weakness of research into eyewitness testimony, for example the research of Loftus and Palmer and Gabbert, is a lack of ecological validity (laboratory setting) and mundane realism (task). Participants are shown film clips of incidents rather than real-life crimes. This means that the task itself lacks mundane realism, because in everyday life witnessing a real crime or a real incident is going to be a lot more stressful for the person watching. However, the research of Loftus has been praised as it has high scientific validity. She was able to manipulate events in a controlled setting, with the IV of the change in verb and the DV of the speed estimate.
​
3. Another weakness of research into eyewitness testimony is in regard to the substitution explanation. Sutherland and Hayne in 2001 found that eyewitness testimony is highly accurate for some aspects of an event. They showed participants a video clip and then asked them misleading questions. They found that participants recall was more accurate for central details of the event compared to the peripheral details. They concluded that the participants attention was probably more focused on the central features of the event as it was important, whereas the peripheral event was less important. They concluded that the central features of an event are largely resistant to misleading information because they are so important.
​
4. Another limitation is that the concept of memory conformity has been questioned. Skagerberg and Wright in 2008 showed participants film clips of a mugging incident. In some cases the muggers hair was dark brown whilst in others their hair was light brown. Participants discussed the clips in pairs after they had seen one of these two different versions. They found that the participants did not report what they had seen in the clips or what they had heard from the co-witnesses but a blend of the two. A common answer to the question of ‘What colour was the mugger's hair?’ was not light brown or dark brown but medium brown. They concluded that the memory itself is distorted through contamination by misleading post event discussion rather than the result of memory conformity.
​
5. Another limitation of research into eyewitness testimony is the use of laboratory studies which often lead to demand characteristics. Zaragoza and McCloskey (1989) have argued that often answers given by participants in laboratory studies are influenced by demand characteristics, as they want to be helpful. Therefore, they may make a guess about a question that they don't really know the answer to.
Anxiety
Anxiety
Anxiety has strong emotional and physical effects, but the research into eyewitness testimony is conflicting. In some cases, the research suggest that anxiety makes eyewitness recall better, whereas in some conditions, it makes recall worse.
​
Anxiety has a negative effect on recall (Weapon Focus)
Johnson and Scott in 1976 conducted a study to look at the effect that the presence of a weapon on eyewitness testimony. They suggested that the presence of a weapon creates anxiety and therefore reduces the accuracy of a witnesses recall for the event. This is weapon focus.
Procedure
They recruited participants for an experiment and asked them to sit in a waiting room. Whilst in the waiting room they were exposed to either a low anxiety condition or a high anxiety condition. In the low anxiety condition, they heard a conversation in the next room, and then saw a man come out of the room with a pen in his hand with grease covering the pen. In the high anxiety condition, participants overheard a heated argument in the next room, followed by the sound of breaking glass and then finally a man coming out to the room holding a knife covered in blood.
Findings and Conclusions
After the event participants had to pick out the man who had walked from the room, either carrying a knife or a pen from a set of 50 photographs. It was found that accuracy in identification was higher when the man was carrying a pen compared to carrying a knife. 49% of the participants who had seen the man carrying a pen was able were able to correctly identify him compared to those who saw the man holding a blood covered knife (33%).
Johnson and Scott explained the findings using the tunnel theory of memory. The tunnel theory of memory proposes that people having an enhanced memory for central events. In the case the weapon can lead to this effect as participants are focused on the weapon and not on the events around the weapon.
​
Anxiety has a positive effect on recall.
Research has suggested that when an individual is in a stressful event, it triggers the fight or flight response which increases alertness. This may mean that memory for that event is enhanced.
Yuille and Cutshall in 1986 conducted a study which linked to a real-life shooting in a gun shop in Vancouver, Canada where a shop owner shot dead a thief. There were 21 witnesses, and 13 of them agreed to take part in Yuille and Cutshall’s study. They were interviewed around 4 to 5 months after the original incident, and then the interviews were compared to the original police interview that was taken at the time of the shooting. Accuracy was determined by the number of details reported in each account. The witnesses were also asked how stressed they felt at the time of the incident, and to give a rating on a seven point scale. They were also asked whether they had suffered from any emotional problems since they had witnessed the event, e.g. insomnia.
Findings and conclusions
It was found that the witnesses were very accurate in their accounts, and that there was very little change, even 4 to 5 months after witnessing the original incident. Some of the peripheral details were less well remembered, but it was found that the participants who reported the highest levels of stress at the time of the incident were actually the most accurate. It was found that the accuracy was around 88% for the high anxiety witnesses compared to 75% for the low anxiety witnesses. They concluded that anxiety does not have a detrimentally effect on the accuracy of eyewitness testimony, and in fact, it can enhance it in a real context.
​
The Yerkes-Dodson Law
One way that this that these contradictory findings has been explained is by using the Yerkes-Dodson inverted U law of arousal. It appears that performance does increase with stress, but only to a certain point after that stress leads to a dramatic decrease in performance. This is sometimes used to explain athletes and their performance in important events, for example, football games.
Deffenbacher in 1983 reviewed 21 studies of eyewitness testimony and noted that there were contradictory findings and used the Yerkes-Dodson law to explain the differences in findings. He suggested that when we watch a crime we become emotionally aroused and this can lead to not only a feeling of anxiety (psychological) but physiological changes in our body because of the fight or flight response. Lower levels of anxiety produce lower levels of recall accuracy, but then memory becomes more accurate as anxiety increases. He suggested that there is an optimal level of anxiety, before it has a detrimental effect on the eyewitness account.
​
Evaluation
1. One strength of research into anxiety is further research which comes from Christianson and Hubinette in 1993. They interviewed 58 witnesses to actual bank robberies in Sweden and found that those who were directly involved (bank workers) who experienced more anxiety were more accurate than the average rate across all witnesses of 75%. However, one limitation of this research is that Christianson and Hubinette only interviewed the participants a few months after the event and therefore it is difficult to control what happened in the intervening time period, e.g. post-event discussion.
​
2. Another strength of the research into anxiety is supporting evidence for the negative effects that anxiety has on the accuracy of recall. A study was conducted by Valentine and Mesout in 2009. They used an objective measure of anxiety (heart rate) to divide participants into low and high anxiety groups. In the study participants went into the horror labyrinth at the London Dungeons where they were subjected to many frightening scare encounters with actors and animated models, etc. At the end of the labyrinth, they were asked to complete questionnaires to assess their level of self reported anxiety. This was then combined with the heart rate monitor measurement in order to divide them into the two groups, high anxiety and low anxiety. The participants were then asked to describe a person that they encountered in the labyrinth (an actor). They found that the participants in the high anxiety condition, had a poorer memory of the actor and made more mistakes when reporting details of them. They also found that only 17% in the high anxiety condition were able to identify the actor in a lineup compared to 75% in those in the low anxiety group. Therefore, they suggested that high anxiety does have a negative effect on immediate eyewitness recall.
​
3. Another limitation of research into anxiety and accuracy of eyewitness testimony comes from a study conducted by Pickle in 1998. In this study Pickle, filmed a video which was set in hair dressing salon. In the salon there were a number of items some of which were unusual, and some of which were highly usual within a hairdressing salon. For example, scissors or a wallet (usual items) whereas a handgun or a raw chicken would be considered to be unusual. They were also classified as causing anxiety or not - scissors and gun causing anxiety, whereas wallet or chicken no anxiety. Pickle suggested that it was surprise rather than anxiety that made eyewitness testimony worse when there was a weapon present, which was supported by the finding that eyewitness testimony was poorer with the unusual objects (chicken and handgun). Therefore concluding that weapon focus is due to unusualness rather than anxiety.