Introduction
Accuracy of eyewitness testimony has been an area of interest to social scientists. According to Bornstein & Zickafoose, (1999) eyewitnesses are not always accurate. Deoxyribonucleic acid (DNA) evidence has attracted many appeals from individuals in the recent past. About 78 percent of those individuals who were acquitted on this basis had originally been convicted based on strong eyewitness testimony. The misinformation effect remains to be a big challenge to witnesses when recalling information about a criminal. The challenge involves the incorporation of misinformation into an individual’s memory after receiving misleading information about an event (stambor, 2006).
Memory can be constructed based on the information given after the fact and also on suggestive questions. Stress and decay might also play a role in one’s information-recalling ability. A memory may also decay over time and stress may reduce the recalling accuracy(Deffenbacher, Bornstein, Penrod, & McGorty, 2004; Payne, Nadel, Allen, Thomas, & Jacobs, 2002).
Methods and Participants
Participants were randomly assigned to one of three levels of stress: high stress, where they were writing a final exam immediately following their participation in the experiment; medium stress, where they were writing a final exam the day following their participation; and low stress, where their participation came two weeks before their final exam. They were shown one of two different versions of a video of a bank robbery and instructed to pay close attention to detail. All versions shared the same beginning scenario, with two individuals entering a bank to rob it.
The first individual (individual #1) was 5’10” tall and of medium build, wearing blue jeans, a black leather jacket, and black tennis shoes. This individual was wearing a ski mask with the holes around the eyes large enough for the color of the skin, which was white or light colored, to be visible. The second individual (individual #2) was 6’2″ and heavyset, wearing black sweat pants, a red jacket, and dark work boots. This individual was wearing a ski mask identical to that of individual #1. The skin around the eyes was dark. No other skin was visible on either individual. Individual #1 walks to the window and hands the teller a note, bringing up the right hand, which was in the pocket, to simulate a gun. It was unknown whether an actual gun was used.
Individual #2 stayed back a step as if keeping watch. After the teller gave money to the robber, the two robbers left the bank, jumped into a car waiting at the curb, and drove away. In version #1, the car was blue. In version #2, the car was green. Following the viewing of the video, each participant met an individual in the waiting room. This individual (actually a confederate of the experimenter) stated that he or she had lost something and had come back to see whether the experimenter found it. The individual begun a conversation about the video and said, “Did you see that blue (or green, depending on the condition) car they were driving?” Half the time, this question was congruent (the color mentioned by the confederate matches the color shown in the video). And half the time, the color was incongruent (the color mentioned by the confederate was different from that visible in the video).
Measures At the end of this conversation, the confederate left, and the participant was asked by the experimenter to complete a short questionnaire to measure his or her memory of the details in the video. This questionnaire consisted of fifteen fill-in-the-blank questions where participants were asked questions, such as “What kind of pants was the individual who asked for the money wearing?” The participant has to write what he or she recalled or “unknown” if uncertain about the response. Each correct answer was given one point, and the points were added to create a continuous measure of recall as the dependent variable. Scores on this measure ranged from 0 to 15. This questionnaire was given again two weeks later, and a third time after one month. Participants were also asked to rate their confidence in their recall of each item on a scale of 1 to 10, with one meaning very little confidence and 10 meaning a great deal of confidence. These items were added to create an overall confidence score, with higher scores indicating higher levels of confidence. Scores on this measure ranged from 15 to 150. The misinformation effect was stored in a variable named “recall color.” Recall color was scored as “Correct” if participants correctly identified the color of the car they had seen in the video and “Incorrect” if they incorrectly identified the color.
Finally, all participants completed a short questionnaire to measure their stress levels. This was necessary to check the stress manipulation and ensure it was effective. Possible scores on this measure ranged from 0 to 25, with higher scores indicating higher levels of stress.
Results
First hypothesis
Null hypothesis: There is no significant relationship between the type of information conveyed (a misinformation effect) and the accuracy of recall about the color of the vehicle.
Alternative hypothesis: There is a significant relationship between the type of information conveyed (a misinformation effect) and the accuracy of recall about the color of the vehicle.
| Type of Information * Recall Color Crosstabulation |
| Count |
| | Recall Color | Total |
| | correct | incorrect | |
| Type of Information | consistent | 27 | 9 | 36 |
| | inconsistent | 15 | 21 | 36 |
| Total | 42 | 30 | 72 |
The results show that out of 36 participants who received consistent information, 27 of them had a correct recall about the color of the vehicle while 9 had an incorrect recall about the color of the vehicle. Also, out of 36 participants who received inconsistent information, 15 of them had a correct recall about the color of the vehicle while 21 had an incorrect recall about the color of the vehicle. This shows that the participants who received consistent information had higher chances of having a correct recall about the color of the vehicle.
| Coefficientsa |
| Model | Unstandardized Coefficients | Standardized Coefficients | t | Sig. |
| | B | Std. Error | Beta | | |
| 1 | (Constant) | .917 | .175 | | 5.227 | .000 |
| | Type of Information | .333 | .111 | .338 | 3.005 | .004 |
| a. Dependent Variable: Recall Color |
The above regression output shows that there is a significant relationship between the two variables since the significant value is less than 5%. That is, 0.004<0.05. Therefore, we reject the null hypothesis and conclude that there is a significant relationship between the type of information conveyed (a misinformation effect) and the accuracy of recall about the color of the vehicle.
Second hypothesis
Hypothesis: Memory decays over time.
Memory decays will be assessed using the recall ability over time.
The results in the above bar chart show that the recall level reduces from recall1 to recall3. This indicates that the recall level decreases with time hence the conclusion that memory decays with time.
Third hypothesis
Null hypothesis: Level of stress does not affect recall
Alternative hypothesis: level of stress affects recall
| level of stress * Recall Color Crosstabulation |
| Count |
| | Recall Color | Total |
| | correct | incorrect | |
| level of stress | low | 15 | 9 | 24 |
| | medium | 12 | 12 | 24 |
| | high | 15 | 9 | 24 |
| Total | 42 | 30 | 72 |
The above output shows that out of 24 participants who had a low level of stress, 15 of them had a correct recall about the color of the vehicle while 9 of them had an incorrect recall about the color of the vehicle. Out of 24 participants who had a medium level of stress, 12 of them had a correct recall about the color of the vehicle while 12 of them had an incorrect recall about the color of the vehicle. Also, out of 24 participants who had a high level of stress, 15 of them had a correct recall about the color of the vehicle while 9 of them had an incorrect recall about the color of the vehicle.
| Coefficientsa |
| Model | Unstandardized Coefficients | Standardized Coefficients | t | Sig. |
| | B | Std. Error | Beta | | |
| 1 | (Constant) | 1.417 | .156 | | 9.087 | .000 |
| | level of stress | .000 | .072 | .000 | .000 | 1.000 |
| a. Dependent Variable: Recall Color |
The regression analysis output above shows that there was no relationship between the two variables since the significant value is greater than 5%. That is 1.000>0.05. Therefore, we fail to reject the null hypothesis at 5% level of significance and conclude that level of stress does not affect the ability to recall.
Fourth hypothesis
Null hypothesis: There is no significant relationship between the level of stress and confidence.
Alternative hypothesis: There is a significant relationship between the level of stress and confidence.
| Coefficientsa |
| Model | Unstandardized Coefficients | Standardized Coefficients | t | Sig. |
| | B | Std. Error | Beta | | |
| 1 | (Constant) | 4.600 | .218 | | 21.107 | .000 |
| | Confidence | -.026 | .002 | -.827 | -12.326 | .000 |
| a. Dependent Variable: level of stress |
The regression equation is; level of stress = 4.600 – 0.026*confidence. The regression equation shows that if confidence increases by one unit, the level of stress decrease by 0.026 units. The regression output also shows that there is a significant relationship between the two variables since the significance value is less than 0.05. That is, 0.000<0.05. Therefore, we reject the null hypothesis at 5% level of significance and conclude that there is a significant relationship between the level of stress and confidence.
Discussion.
According to the results from this research, an individual’s accuracy of recall depends on the type of information conveyed ( a misinformation effect). A person is more likely to have a correct recall if the information conveyed to him/her is consistent. A person conveyed with inconsistent information is more likely to have an incorrect recall.
The results from this researcher also show that a person’s memory decays with time. This is because the level of recall was found to decrease as time goes by. Therefore, time is a determinant of the ability to recall. These findings are similar to those of Shapiro & Penrod (1986) that stated that memory may decay over time.
The findings of this research also indicate that a person’s level of stress does not affect his/her ability to recall. These findings are in contrast with those of Deffenbacher, Bornstein, Penrod, & McGorty in 2004 and Payne, Nadel, Allen, Thomas, & Jacobs in 2002 that stated that as stress increases, the accuracy of recall decreases.
The researcher also found that there is a significant relationship between a person’s level of stress and his/her confidence. It also showed that confidence decreases as stress increases.
Direction for future research.
· Research should be carried to determine whether there is a significant relationship between age and the ability to recall.
· Research should be carried to determine whether there is a significant relationship between gender and the ability to recall.
References
Bornstein, G. HY., & Zickafoose, D. J. (1999). “I know I know it, I know I saw it”: The stability of the confidence-accuracy relationship across domains. Journal of Experimental Psychology: Applied, 5, 76–88.
Deffenbacher, K. A., Bornstein, B. H., Penrod, S. D., & McGorty, E. K. (2004). A meta-analytic review of the effects of high stress on eyewitness memory. Law and Human Behavior, 28(6), 697–706.
Myers, D. G. (2008). Social psychology (9th ed.). New York, NY: McGraw Hill.
Payne, J. D., Nadel, L., Allen, J. J. B., Thomas, K. G. F., & Jacobs, W. J. (2002). The effects of experimentally induced stress on false recognition. Memory, 10(1), 1–6.
Shapiro, P. N., & Penrod, S. (1986). Meta-analysis of facial identification studies. Psychological Bulletin, 100, 139–156.
Stambor, Z. (2006). How reliable is eyewitness testimony? APA Monitor, 37(4), 26–27.
Bar Chart
Recall1 Recall2 Recall3 542 473 369
Time
Recall Level
