Recent Research: Effects of Stress on Police and Citizen Eyewitness Recall
Zimmerman, Laura A., Canadian Journal of Police and Security Services
Stanny, C. J. & Johnson, T. C. (2000). Effects of stress induced by a simulated shooting on recall by police and citizen witnesses. American Journal of Psychology, 113(3), 359-386.
Research indicates that stress experienced by eyewitnesses during criminal events may be detrimental to later recollection of those events (c.f. Narby, Cutler, & Penrod, 1996), but it may be argued that these levels of stress may vary depending on who the witness is. Police officers make up one population of eyewitnesses who frequently encounter criminal activity and, as eyewitnesses they perform the same tasks as citizens. During the investigation, both police and citizen eyewitness describe the event and perpetrator for police records. Once the case goes to trial, witnesses must testify in court about what they saw. Whether police officers are under more or less stress than citizens when experiencing criminal events, remains unclear. Police officers may experience less stress because they receive training and experience for dealing with crime events which allows them to remain calm and to apply constructive decision strategies to dangerous situations. On the other hand, police may experience more stress than the average person witnessing a crime. While citizens need only to protect themselves and avoid crime events, police must actively confront the situation, protect bystanders and themselves, and make spur-of-the-moment decisions about the level of force necessary to defuse the situation. It is these factors that may lead to increased levels of stress.
Although it is commonly thought that the ability of police officers to recall witnessed criminal activity is greater than the ability of civilians (Yarmey, 1986), very little research has been conducted comparing the amount of details correctly recalled by police and citizens. Results of these studies generally indicate that police are sometimes more accurate in their descriptive detail of criminal perpetrators and crime scene activity. These results, however, have been inconsistent (Zimmerman, 2003). If police do recall events better than citizens, and stress is related to a decrease in recall, this would seem to indicate that police experience less stress in crime situations. Very little empirical research has been conducted which directly manipulates levels of police stress in order to determine the effects on police recall accuracy. One such study by Yuille, Davies, Gibling, Marxsen and Porter (1994) found that police trainees in a stressful condition recalled fewer correct details than police trainees in a non-stressful condition. However, this study did not compare police performance to citizen performance. Because the comparison between police and citizens has rarely been made, it has remained unclear if there are differences in police and citizen accuracy under different levels of stress.
As such, two studies conducted by Stanny and Johnson (2000) sought to clarify whether: (a) officers in stressful situations experience a decrease in later recall, and (b) differences in recall ability exist between police and civilians and if those differences are related to stress levels. To assess these questions, the authors used simulated scenarios involving violent crimes which sometimes escalated into shooting incidences. Participants performed in the role of either active participants or as bystanders witnessing the crime. Prior studies have shown that active participants (i.e., victims and police) give more accurate descriptions than bystanders (Christianson & Hubinette, 1993; Hosch & Bothwell, 1990). One concern raised in these studies however was that active participants and bystanders witness the event from different points of view, with active witnesses having more access to information. After participants in the current study witnessed the event, they attempted to recall details of the scenario by completing a questionnaire concerning the incident.
In Study 1, Stanny and Johnson, assigned 40 police officers to the role of either active participant or bystander in order to assess differences in recall accuracy among participants who had similar training and experience. Two scenarios were presented to all participants. One scenario depicted a domestic disturbance in which a gun was drawn by the perpetrator and immediately placed on the ground (i.e., no-shoot scenario). The other scenario depicted an attempted abduction. During this scenario the perpetrator pulled a gun, aimed, and fired once (i.e., shoot scenario). The scenarios were presented on the Firearms Training System, a computerized, interactive, police training system designed to create a realistic experience (c.f. Stanny & Johnson). Participants moved about the scene, issued orders, drew, and fired their weapons if deemed necessary. Bystanders performed the role of witnesses who viewed the event from the same perspective as the active participants, but did not become involved within the scene. After viewing the event, participants completed a 30-item questionnaire containing questions about the victim (9 questions), perpetrator (9 questions), weapons (6 questions), actions (2 questions), and other details (4 questions).
Results of Study 1 revealed that participants recalled a larger proportion of correct detail from the no-shoot scenario. In the shoot scenario, however, memory for details about the perpetrator was significantly worse than memory for all other details. Memory for the perpetrator was also significantly worse in the shoot condition compared to the no-shoot condition. The hypothesis that active engagement in an event might influence the accuracy of eyewitness memory was not supported. Both active participants and bystanders displayed decreased recall ability in the shoot condition. Therefore, active engagement did not produce a greater decline in recall compared to bystanders in either the shoot or the no-shoot condition. The authors concluded from Study 1 that police officer recall ability is influenced by the presumed emotionality and heightened arousal associated with shootings, regardless of their participation level.
Although informative, Study 1 did not directly compare police to citizens. The performance of police officers may still be greater than untrained, inexperienced citizens. The purpose of Study 2, therefore, was to compare recall ability of police eyewitnesses to citizen eyewitnesses. In this study 16 police officers and 13 university students witnessed multiple scenarios, one with a shooting and one without a shooting. Police officers were all assigned to the active participant condition, while the university students were assigned to the role of bystander. Both the domestic disturbance and attempted abduction scenarios were manipulated to show either a shooting or a non-shooting ending. Measurements of stress were recorded by applying electrodes to each participant and measuring their physiological arousal during the event. Stress was also assessed with self-report questionnaires which directly asked participants about the stress they experienced and about perceived levels of violence.
Results of Study 2 indicate that physiological stress was significantly elevated in the shoot condition compared to the no-shoot condition for both police and citizens. This finding suggests that police and citizens experience similar levels of stress even though only the police engaged in a shooting. Overall, police reported experiencing more stress than citizens but all participants in the shooting condition reported more stress than those in the no-shoot condition. Similar to Study 1, all participants in the no-shoot condition recalled a greater proportion of correct details compared to those in the shoot condition. Police and citizens did not differ in recall accuracy in either the shoot or no-shoot condition. Further analysis revealed that in the shoot condition the weapon was remembered better than all the other details.
The authors demonstrated in Study 2 that police experienced an increase in stress similar to citizens and that stress significantly increased when a shooting occurred. Based on this finding, they concluded that increased stress is associated with a decrease in eyewitness recall accuracy for both police and citizens. The authors also suggest that effects of increased stress may mask any heightened recall ability that police officers might possess. That is, training and experience may enable officers to achieve levels of recall accuracy similar to those reported here when they are in highly stressful situations. Since citizens did not perform in the active role, it is unknown if their accuracy would be affected. In Study 1, the mean accuracy rates of active and bystander police-participants were not reported, so it is unknown if police-bystander recall ability matched citizen-bystander ability.
Although this article provides some interesting findings regarding the issue of stress, experience and recall, there are a number of limitations. To begin with, since the number of participants in these studies was small there may not have been enough power to detect differences in these comparisons even if differences really do exist in the population. Reported effect sizes were generally small, indicating that a larger number of participants may be necessary to detect an effect. That is, expected differences may not have been detectable because the sample was too small or because there really is no effect. This question may be answered by replicating the study with a larger sample. As suggested by the authors, it would also be useful to gather more data before making definitive conclusions about the performance ability of police compared to civilians.
Another limitation has to do with the recall method used in the study. A 30-item cued recall questionnaire was administered with a limited range of questions asked for each category (i.e., victim, perpetrator, action). For instance, only two action detail questions were asked and both had to do with the police officers' actions, not perpetrator actions. Research has shown that using free recall methods provide more accurate and detailed information (Fisher, 1995). As such, using free recall may have enabled participants to report the full breadth of what they remembered and may have revealed differences between police and citizen eyewitnesses. Furthermore, the questions asked seem representative of questions commonly asked by police, who may indeed be more experienced in remembering these details. If asked about details that are not anticipated, recall may decrease.
Despite the above limitations, this study has some very definite advantages over previous work in the area. First, by employing an interactive video simulation of a crime event, this research effectively manipulated stress in a way that other studies measuring the effects of stress on recall have not done. Inducing stress in study participants by having them experience a real-life violent crime has obvious ethical limitations. The use of this police training module provided participants with a realistic simulation that successfully elevated stress levels. This study is also one of very few studies to employ physiological measures of stress instead of relying solely on self-report measures. As revealed by these findings, self-report measures do not always accurately reflect implicit measures. Police reported feeling significantly more stress than citizens yet their physiological stress levels were similar to citizens.
This article provides important information for police officers who often perform the role of eyewitnesses when reporting crimes. Not only is it important to understand how police are affected when actively engaging in crime events, but also how their later reports of the events are impacted. This has implications not only for the initial search and apprehension of suspects but also for when police officers testify in the courtroom. First, police may report being under more stress during criminal events, especially when weapons are present and fired, but this study indicates that this stress does not necessarily produce a decline in memory beyond that of citizens. Second, trained and experienced officers may not be any more accurate than citizens when recalling details of an event. This indicates that the same diligence that is employed when questioning citizen witnesses should be employed with police witnesses.
Christianson, S.A. & Hubinette, B. (1993). Hands up! A study of witnesses' emotional reactions and memories associated with bank robberies. Applied Cognitive Psychology, 7, 365-379.
Fisher, R. P. (1995). Interviewing victims and witnesses of crime. Psychology, Public Policy, and Law, 1, 732-764.
Hosch, H. M. & Bothwell, R. K. (1990). Arousal, description and identification accuracy of victims and bystanders. Journal of Social Behavior and Personality, 5, 481-488.
Narby, D. J., Cutler, B. L., & Penrod, S. D. (1996). The effects of witness, target, and situational factors on eyewitness identifications. In S. L. Sporer, R. S. Malpass, & G. Kohnken (Eds.), Psychological issues in eyewitness identification (pp. 23-52). Mahwah, NJ: Erlbaum.
Yarmey, A. D. (1986). Perceived expertness and credibility of police officers as eyewitnesses. Canadian Police College Journal, 10, 31-51.
Yuille, J. C., Davies, G., Gibling, F., Marxsen, D., & Porter, S. (1994). Eyewitness memory of police trainees for realistic role plays. Journal of Applied Psychology, 79, 931-936.
Zimmerman, L. A. (2003, October). A systematic review of police officers as eyewitnesses: Do police describe and identify criminal perpetrators better than civilians? Invited paper presented at the 3rd Annual Conference on Eyewitness Identification and Memory, Florida International University, Miami FL.
Laura A. Zimmerman *
Department of Psychology
University of Texas at El Paso
* Comments concerning this paper can be addressed to: Laura A. Zimmerman, Department of Psychology, University of Texas at El Paso, El Paso, Texas 79968 (email: email@example.com).…
Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Recent Research: Effects of Stress on Police and Citizen Eyewitness Recall. Contributors: Zimmerman, Laura A. - Author. Journal title: Canadian Journal of Police and Security Services. Volume: 1. Issue: 4 Publication date: December 2003. Page number: 377. © 2006 Meritus Solutions, Inc. COPYRIGHT 2003 Gale Group.
This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.