However, several real-life studies have been conducted into the effects of leading questions on recall, such as Yuille and Cutshall (1986), who interviewed witnesses of shootings, and Christianson and Hubinette, who did the same with bystanders and victims of bank robberies. Both of these found that central detail shown that recall of central facts is unaffected and that only minor details are forgotten or distorted.
At the time of witnessing a crime, people are usually in a state of high emotion and evidence shows this has an effect on later recall. ‘Flashbulb memories’ are memories of specific important events of which recall of detail remains surprisingly good a long time after the event has taken place and are a popular theory first proposed by Brown and Kulik. Johnson and Scott conducted a laboratory experiment in which participants were subjected to either a high or low stress condition. The high stress condition (witnessing a man coming out of a room holding a blood-covered letter opener) was assumed to cause sufficient emotional arousal for a flashbulb memory of the event and indeed found that some participants from this condition had better recall when asked to identify the man, however this was by no means true for all of them. An opposing theory of the effect of emotional arousal on recall is that of repression, when a person locks potentially disturbing or painful memories into the subconscious so they are not accessible, explaining why many victims of crime remember nothing of the event.
Studies of the effect of emotional arousal often use retrospective data so the memories may have already been distorted – there is no way of knowing. Many others take place in laboratories, eliminating this problem but these studies lack ecological validity and may yield demand characteristics.
In an attempt to solve both these problems two experiments will be conducted, one in a field setting and one a laboratory to find out if ecological validity, i.e. the high emotional state that witnesses are in, has an effect on recall. Due to the fact that there is evidence for both flashbulb memories and repression, a two-tailed hypothesis will be more appropriate as preliminary research has shown that although ecological validity is likely to have some effect, it is unclear as to what this will be (whether it will increase or decrease recall.)
This research will be fairly similar in set-up to that of Loftus, with people watching a scene take place and then being asked questions about it. Her work on leading questions will also be brought in to determine if people’s susceptibility to these is different in the two situations by making one of the ten questions asked leading, although this will not be tested statistically and is simply an additional point of interest.
The aim of this study is to determine whether or not ecological validity, i.e. the high emotional state that people are in at the time of witnessing a crime, has an effect on recall in eyewitness testimony.
The experimental hypothesis is that ecological validity will have a significant effect on eyewitness recall.
The null hypothesis is that ecological validity will have no significant effect on eyewitness recall.
Method
Design: Two different methods were used for the study; the first part was a field experiment and the second a laboratory experiment. This was because the aim was to compare different levels of ecological validity, which these two situations provided as one was highly artificial and one more realistic. To avoid participants becoming aware of the nature of the study, two independent groups were used. If this had not been done, and the same sample used for both parts, participants would have remembered the questions and known what to look for, and they may have guessed the purpose of the experiment.
The independent variable was the environment, and the dependant was whether participants were correct in their answers. A confounding variable which was not considered was that participants may have been able to talk about what had been seen before they were asked the questions; this would have distorted results and if the experiment was repeated, they would have been given questionnaires to all fill out simultaneously in order to avoid this. Another confounding variable would have been participants partaking in both experiments; this was avoiding by ensuring two separate classes were chosen.
Participants: For each situation, an opportunity sample of 10 females aged 12 and 13 was used; all were from a selective grammar school, ensuring they were of similar intelligence. They were chosen purely for the time at which they were available and were already divided into groups as they came from two classes.
Materials: For the first experiment, we used a list of ten questions (see appendix 1) and an answering sheet consisting of a participant number and a tick/cross system. For the second, the same question list and answer sheet were used, however a pre-recorded version of a situation identical to that the first group of participants witnessed was projected onto a screen.
Procedure:
- After finding a suitable teacher, we gave a drama student a rough script of an argument, including key phrases, and allowed them to rehearse
- We then used an empty classroom to videotape the scene, in which an angry student bursts in and shouts about the grade she has been predicted on her UCAS form. Key phrases included the teacher calling her a “pikey”, as we felt this would be remembered clearly by the participants as it was unexpected. The argument lasted approximately 2-2 ½ minutes.
- We then staged the same argument for real in front of a pre-decided class of year 8 students who were unaware that the situation was artificial.
- After our drama student stormed out of the room, we entered and explained that they had taken part in a psychology experiment, then after asking them not to discuss what happened took them individually outside for questioning.
- Each participant was asked the same ten questions; for example ‘What was the last thing the teacher said?’ One of these was a leading question. Responses were filled in on the answer sheet.
- All participants were fully debriefed (see appendix 1) and all questions answered, before permission to use their results was obtained.
- Another year 8 class were informed they were taking part in an experiment concerning eyewitness testimony (but not about the leading question) and then shown the pre-recorded video.
- Again, participants were taken outside individually, asked the same questions and then debriefed.
Ethics: For the field experiment, participants were not made aware of their right to withdraw at first, as doing so would have alerted them to the fact they were in an experimental situation and so would have had a significant effect on our findings. Both the field and laboratory experiments raised the issue of fully informed consent; the first because participants were not made aware of the fact they were part of an experiment and the second because they were not told there were leading questions. However, if they had known of this the validity would have been compromised due to demand characteristics, i.e. they would have known what was expected of them and so either tried to do this, or do the opposite. Both would have distorted results. Informing the participants from the field experiment of the situation would have lowered ecological validity and emotional arousal, meaning an accurate comparison between the two situations would not have been possible. Although the individuals themselves had not consented, prior general consent had been obtained from their parents, which gives some indication that participants would not have objected. This is backed up by the fact that, when given the chance to withdraw their results, none did so. As fully informed consent was an unavoidable part of the experiment, steps were taken to minimize the risk to participants, for example fully debriefing them afterwards to ensure they knew what they had been involved in, and that there was no physical or psychological damage.
Discussion
As shown in the results section, ecological validity has not been found to have any significant effect of accurate recall in eyewitness testimony, meaning that the experimental hypothesis had to be rejected in favour of the null. Initially the fact that the difference between the two situations was so small was surprising (average of 5.9 correct answers for high ecological validity compared to 5.3 in the low ecological validity condition), as was the fact that out of 20 participants, only one person was able to answer the leading question correctly; she was from the low validity group and so would have been studying the scene more closely.
Although the null hypothesis has to be accepted, I feel that although the results were not statistically significant, they do still provide some support for the evidence looked at in the introductory section.
The findings were in keeping with Loftus and Palmer’s study into the effects of leading questions of eyewitness testimony, as all but one participant failed to answer the leading question correctly. If the question had been “Did she put her bag down” as opposed to “Did she put her bag down on her left or her right”, it is likely that the participants would have thought more closely about whether or not she actually did, rather than automatically assume that she did and so guess at an answer. As this shows that the phrasing of the question affects how a witness answers it, this shows support for Loftus and Palmer. Due to Loftus’ research into the ‘weapons focus’, the questions put to the participants concerned both central and peripheral detail, for example asking what grade the student got when she repeated it many times and this was the main focus of the episode; consequently 90% of participants were able to recall correctly. When asking less central questions, recall decreased e.g. whether she was wearing a watch (35% correct) and so I believe my research also supports this.
My research has several methodological limitations, the first of which being that the field experiment would not have created sufficient emotional arousal to produce a flashbulb memory; however to do so on young participants unaware they were taking part in an experiment would have been highly unethical. It would have been difficult to have used a sample in which people of both genders of all ages and backgrounds were represented, but the sample for this study was extremely limited (grammar school girls aged 12-13) and so the findings cannot be generalized to the wider population. However, this does reduce the likelihood of individual differences between the two groups. Using more questions may have been more useful, but this could have led to an overall decrease in recall as participants may have found remembering many small details as opposed to a few challenging, however using more participants would have been a great improvement.
A major limitation of the study is that while participants were being questioned individually, the rest may have been able to talk about what they saw, and when the questioned participant returned she may have informed the others of what questions were asked. This goes some way towards explaining why similar answers were obtained for several questions, although this could also be due to the fact that the sample was so limited and therefore all participants had the same ideas and schemas. If the experiment was to be repeated, I would use more participants from a wider sample and ask a few more questions, with equal focus on visual and acoustic recall. I would also use a written questionnaire to obtain their answers, as this would ensure all participants answered at the same time and were unable to confer. I believe these changes would improve the experiment and provide more conclusive results.
Further research could be carried out into the difference in recall of visual and acoustic details when in a state of high emotional arousal. An example of a field experiment testing this could be staging a fight between two people, questioning witnesses about equal amounts of both visual and acoustic aspects and then comparing the results to find out which is recalled better when emotionally aroused. Important research could also be carried out into effect of emotional arousal on how an individual responds to a leading question as this could prove very useful, for example if there were two witnesses to a crime and one was fairly calm while the other was not, which would be most likely to give more accurate evidence.
This experiment has implications in the field of eyewitness testimony – it suggests that reliability is not affected by differing levels of emotional arousal, which can prove reliability of witnesses in court and give more weight to their testimonies, which is important in terms of criminal cases. However, it also shows that witnesses are easily led by something as simple as the wording of a question and also that when someone fails to recall something they are liable to invent an answer, which somewhat reduces their reliability. This implies that more open-ended questions should be used so there is less scope for misleading.
Although this may be due in part to the methodological limitations (small sample, few questions, possible conferring), the research has not shown differing levels of emotional arousal to have any significant effect on recall.
Appendix 1
Questions Asked: (Including answers)
- Was Becky wearing trousers or a skirt? (Trousers)
- What colour top was Becky wearing? (Red)
- Was Becky wearing a watch? (No)
- What was Becky holding in her hand? (A sheet of paper/UCAS form)
- What grade had Becky’s application been given? (E)
- What colour hair did Becky have? (Blonde)
- What was the first thing Becky said? (‘What is this?)
- What was the last thing the teacher said? (‘Get out of my class room’)
- Did Becky put her bag down on her left or her right? (Neither - she kept it on her shoulder)
- Which university did Becky want to go to? (Exeter)
Condition 1 – High ecological validity:
Standardised de-brief used for the participants in condition 1:
“ Thank you for taking part in our psychology experiment. We are researching into the effects of ecological validity, which is the “realness”, of a situation on eyewitness testimony. We would be grateful if you could answer a set of 10 questions, based on the scene you have just witnessed, but you are under no obligation to do so. At any point during the questioning you can withdraw, and you can also ask us not to use your results in our findings.”
Condition 2 – Low ecological validity:
Introduction used for the participants in condition 2:
“We are psychology students studying the effect of ecological validity (“realness”) on eyewitness testimony. We would ask you to study the following film clip and then answer a set of 10 questions based on what you have seen. You have the right to withdraw at any time during the experiment and can also withdraw your answers from our results section if you wish.”
(Participants in both conditions were given the correct answers in debriefing, and informed of the leading question.)
Script of argument
Student: What’s this?
Teacher: I don’t know, what is it?
S: How can you have predicted me an "E" on my UCAS reference? Don’t you understand how important this is? I don’t understand why you would do this, this is completely unfair! Do you really think Exeter will accept me with this reference? This is going to ruin all my chances.
T: Becky, this outburst is not acceptable. Just calm down. Why don’t you have a seat and we’ll talk about this rationally?
S: No! I will not calm down until you admit this prediction is just a way of venting your own personal vendetta towards me! This mark is completely unjustified and we both know it.
T: Unjustified? It happens to be my opinion that you will end up with an "E" in my course. Maybe you would have had a better grade if you hadn’t spent all term chatting to your mates and messing around, you’ve missed at least 3 deadlines and missed several lessons.
S: What? I’ve missed one lesson when I was absent, and made all but one of the deadlines! I knew this was because you’ve got it in for me. You have done ever since I came to your class-always grading me below the other students, marking me as absent when you can see that I’m present, I’ve seen you do it! This is completely unprofessional and…
T: Listen, don’t try to make out that this is personal; I simply don’t believe that you will do well in the subject, and I believe I graded you fairly. Maybe you should set your sights at a lower university, face it, you’re not really Exeter material anyway, are you?
S: If you don’t change this mark, I’m taking this higher, and I will get results.
T: Are you trying to threaten me? I don’t have to put up with this, Becky. Get out of my classroom.