Method
Participants
Participants included 20 students at Swansea University with zero of them studying psychology, which included 12 males and 8 females. The average age of the participants was 22.75 years, with a range of 19-28 years. All of the participants remained anonymous to the researchers and all of them reported normal vision or corrected-to-normal vision. All of them had not done the same kind of experiments before and participants were consented to participate in this study by signing a consent form.
Stimuli
The stimuli were slides that depicted the event before, during and after a handbag-snatching criminal incident. Each slide was presented to participants onto a screen for 3 seconds, by using Microsoft PowerPoint. The slides were from Catrin Jones and Cathryn Lloyd-Davies. Apart from presenting the video to participants, there are two types of questionnaires will be given to participants as well. Phase 1 questionnaires consisted of central and peripheral questions, for example, one of the central questions is “What colour is the main garage door?”, and one of the peripheral questions is “What is all over the floor?” The second part of questionnaires (phase 2) was also consisted both central and peripheral type of questions, for example, the central question is: “Where is the motorbike logo on the assailant’s clothing?” and the peripheral question is: “What colour was the closest car to the left of the screen?” Distracter tasks were included in the experiment and were used in-between viewing the event and Phase 1 questionnaire and in-between phase 1 and phase 2 questionnaires.
Apparatus
All trial presentations were controlled by Microsoft PowerPoint and were presented via laptop. The screen image measured approximately 0.30m x 0.20m.
Design
The study was based on a mixed design. Participants were divided into two groups and take parts in two different questionnaires based on the same video clip. The within-subject factor will be focus of the question and will have two levels: Central and peripheral, all participants were required to finish both of the levels. The between-subject factor will be the group and will have two levels: half of them were in control group and half of them were in misinformation group. The first part of questionnaires contained 12 questions and the second part of questionnaires contained 6 questions. Each part of the questionnaires contained central and peripheral types. The dependent measure will be recall accuracy calculated in percentage. Distracter tasks and questionnaires were allocated to participants by the counterbalancing method.
Procedure
Participants were divided into two groups. The first group was misleading group and the second group is control group. Participants were allocated to conditions randomly by the experimenters.
Several ethical considerations have been concerned. First, participants have to understand what they are going to do in the study; they have also been told that the questionnaire is about an eyewitness testimony. Second, they have to understand that they can withdrawal anytime if they want. Third, a consent form will be given to participants before they started, they have to sign it in order to confirm they understand what they are required to do and they can quit the experiment at anytime if they want. Since the video clip is about a criminal event, therefore there are foreseeable potential risks to participants. It may recall participants’ similar experiences in the past which make them feel uncomfortable and also the knife in the video may affect participants’ emotions. Regarding these potential risks, contact details of student counselling service is provided in the debriefing form. Participants were all welcome to contact them for any further queries. Finally all data will be kept confidential and participants’ results will not be recognized. This experiment contains the same risks to both experimenters and participants as mentioned above.
This experiment took place in a private environment, which was a room provided by the experimenters and only one participant could take part in the experiment each time. The participant received an information sheet which informed them of the nature of the study. Before playing the video clip to participants, they were required to sign the consent form in order to make sure they understood how the experiment is going to run and they could withdraw from the experiment at anytime. The participant’s task was to finish a 3 minutes distracter sheet followed by phase 1 questionnaires, then a second distracter sheet will be given to participants again and again, they will have 3 minutes to finish it. After that, phase 2 questionnaires were given to them before they received the debriefing form. Only correct answers will be recorded for the results.
Results
The means and standard deviations of the correct recall for central and peripheral events for each of the two groups in Phase 1 are presented in Table 1 and the means and standard deviations of the correct recall for central and peripheral events for each of the two groups in Phase 2 are presented in Table 2.
Table 1: Mean percentage (Standard Deviations) of the correct recall for central and peripheral information for the Control and Misleading group in Phase 1.
Table 2: Mean percentage (Standard Deviations) of the correct recall for central and peripheral information for the Control and Misleading group in Phase 2.
Phase 1 Analysis:
Phase 1 scores for central events is 51.70, R-squared =0.08. The relationship is linear.
Mean percentage accuracy scores were entered into a 2 (Group: Control vs. Suggestibility) x 2 (Central & Peripheral) ANOVA. The effect of Group for misinformation was significant (F (1, 9) = 40.096, p =0.000). The effect of Group for Control was significant (F (1,9) = 9.45, p=0.001) The main effect of Central was not significant (F (1, 18) = 3.461, p = 0.079). The main effect of Peripheral was not significant (F (1, 18) = 0.758, p = 0.396). The interaction between the Group and Focus was significant (F (1, 18) = 6.010, p = 0.025). Therefore, the two groups did not differ significantly from each other in memory recall of details in the event. However, it shows that central events were remembered better than peripheral events.
Phase 1 Scatter-plot
Phase 2 Analysis:
Phase 2 scores for central events is 62.55, R-squared =0.03. The relationship is linear.
Mean percentage accuracy scores were entered into a 2 (Group: Control vs. Suggestibility) x 2 (Central & Peripheral) ANOVA. The effect of Misinformation Group was significant (F (1, 9) = 14.879, p = 0.004). The effect of Control Group was not significant (F (1, 9) = 0.130, p = 0.726). The main effect of Central was not significant (F (1, 18) = 1.8, p = 0.196). The main effect of Peripheral was significant (F (1, 18) = 15.784, p = 0.001). The interaction between the Group and Focus was significant (F (1, 18) = 4.882, p = 0.040). Therefore, the two groups were significantly different from each other in memory recall of details in the event. The memory recall for central events was not significantly different between the Control and the Suggestibility group (t (1,18) = 1.342, p =0.196). However, memory for peripheral events was significantly different between the two Groups (t (1,18x) =3.973x, p =0.001). Therefore, it can be concluded that misleading post-event information affects peripheral event memory more than the central events memory.
Discussion
The aim of the present experiment was to examine the effect of misleading information (misinformation) on recall of central and peripheral details of a crime event and also, how much it affects and linked to each other. Previous literature has suggested that misinformation effect occurs and would lead to a poorer recall. (Henry et al, 1995) The prediction is that misinformation will affect memory recall and lead to less accurate of recalling event details and there will be a different in remembering central event and peripheral event. In accordance with previous literature, we found that participants recalled more details on the central event compared with the peripheral event and misinformation effect occurs to participants. Our results accepted our hypothesis with a significant difference in recall of events between the misinformation and the control groups.
However, there are several problems which can be done better in the present experiment. First of all, as the questionnaires were set as a Multiple Choice questions, it affected the accuracy of the results. It is because participants could get the answers correct by guessing, even they were not sure for the answers nor they were totally unknown about the details of questions, they could just guess and circle one of the answers provided. Because of this reason, the results were not able to show the accuracy on how misinformation was affecting memory recall. It is because the poor results of recalling could be explained by the higher probability of guessing the wrong answers from the questions. To improve this, Multiple Choice Questions should not be included in the experiment but instead, choose to use free recall test. By giving some opening questions (which also included misinformation and repeated measure) to participants, then required participants to write down what they could remember form the event.
Also, because of the limitations of the experiments, the numbers of participants are not enough to carry out an accuracy result as there are only 20 participants taking part in this experiment. In order to have a more accurate results or a clearer finding on the degree of misinformation affect memory recall, more participants are needed and should be selected from different groups of people, for example different age-group. It is because there are huge individual differences among people (which will be discussed later) and especially within different age group. In the current study, the age range of participants is from 19-28 and they are all students.
Regarding the individual differences problems, there are several points which were made. Firstly, there are researches shown that people in different anxiety levels will have significant differences on recalling event. (Dobson et al, 1993) Those subjects with a higher anxiety levels were less accurate on an eyewitness task than those who scored low anxiety levels on the Test Anxiety Scale. (TAS; Sarason, 1972) There is also some suggestions that stress will decrease memory performance, because each participant was living under different stress conditions, it is highly depends on their own life. Therefore before we start the experiment, participants should have been scored on the Test Anxiety Scale so that we could filter some irrelevant results and a more accuracy results will be found. Second, personality affects memory functioning as well. Siegel and Loftus (1978) found that anxious and self-preoccupied subjects were poor witnesses.
Last but not least, there are some researches shown that weapon effect appeared to be affecting human recall in crime event. (Johnson & Scott, 1976; Loftus et al, 1987) Weapon effect is that when a weapon is present in a criminal event, the victim or witness will look at a weapon more than at other things, therefore their memory for other things will be less accurate. It is believed that victim or witness may search for possible escape route rather than encoding relevant information about the criminal. (Anne M., Gunther K., 1989) As a weapon was included in this experiment, weapon effect could be explained as one of the reasons why participants’ recalls were poor. It affected the aim of experiment as researchers could not analyse an accurate results on how misinformation was affecting memory recall. Researchers were not sure whether the poor results of participants’ recall were because of the misinformation or the weapon effect. Therefore it is suggested that in the future experiment, weapon should not appear in the video slide in order to have a more accurate results on how misinformation affecting human memory recall.
Basically the results are as predicted. The methods and procedures went appropriately and the experiment tested what the experimenter intended to test. Although there were several problems in the current study, problems were found and hence there is a foreseeable improvement in the future experiments.
In sum, this experiment found that misinformation will lead to a poorer recall and less accurate details about a criminal event and central event details will be remembered better compared to peripheral details. Also, two group of participants presented a different accuracy in recalling two different types of details (central and peripheral) Finally, we can confirm what Henry (1995) suggested that misinformation effect occurs in memory recall is correct.
Reference
Anne M., Gunther K., (1989). Simulating the “weapon effect’’. Law and Human Behaviour, 13, 397-408
Dobson, M, & Markham, R (1993). Individual differences in anxiety level and eyewitness memory. The Journal of General Psychology. 119, 343-350.
Henry , R., Derek , J., & Kathleen, B. (1995). Misinformation Effects in Recall: Creating False Memories through Repeated Retrieval . Journal of Memory and Language. 35, 300.
Johnson. C., & Scott B. (1976). Eyewitness testimony and suspect identification as a function of arousal, sex of witness, and scheduling of interrogation. Paper presented at the American Psychological Association meeting.
Loftus. E. F., Loftus, G. R. & Messo. J. (1987). Some facts about “weapon focus”. Law and Human Behaviour, 11, 55-62.
Sarason, I. G. (1972). Experimental approaches to test anxiety: Attention and the uses of information. In C. D. Spielberger (Ed.), Anxiety: Current trends in theory and research, Volume 2 (pp. 381-403). New York: Academic Press.
Shiffrin, R.M., & Steyvers, M. (1998). The effectiveness of retrieval from memory.
In M. Oaksford & N. Chater (Eds.), Rational Models of Cognition, pp. 73-95 Oxford, U.K.: Oxford University Press.
Malpass, R., & Kochnken, G. (1995). Psychological issues in eyewitness identification..Psychology Press.
Appendix
Descriptive Statistics
Table 1: Phase 1 Control Group
Descriptive Statistics
Table 2: Phase 1 Misinformation Group
Descriptive Statistics
Table 3: Phase 2 Control Group
Descriptive Statistics
Table 4: Phase 2 Misinformation Group
Tests of Within-Subjects Effects
Measure: MEASURE_1
Table 5: Phase 1 interaction
Tests of Within-Subjects Effects
Measure: MEASURE_1
Table 6: Phase 1 Mislead
Tests of Within-Subjects Effects
Table 7: Phase 1 Control
ANOVA
Table 8: Phase 1 Focus
Independent Samples Test