|
Article Excerpt
Stress in police work can result in distress for the officers (e.g. Anshel, Robertson, & Caputi, 1997; Hart, Wearing, & Headey, 1995; Kaufmann & Beehr, 1989), but its impact on the officer's work performance is less well established. The present study examined police officers' work-related memories for a potentially stressful event, memories that are directly related to effective job performance. The most common organizational contexts in which researchers explicitly study memory are performance appraisals (e.g. DeNisi & Peters, 1996; Lance, Woehr, & Fiscaro, 1991; Sanchez & De La Torre, 1996; Woehr & Lance, 1991) and training (e.g. Warr, Allan, & Birdi, 1999). The focus of the present study is the accuracy of police officers' reports of a dangerous incident involving shooting. The accuracy and reporting of such incidents impact their job performance and ultimately the effectiveness of both the police organizations and the larger justice system through officers' subsequent testimony in courts.
Although research on a wide variety of predictors of performance in police settings is common (e.g. Armeli, Eisenberger, Fasolo, & Lynch, 1998; Cortina, Doherty, Schmitt, Kaufman, & Smith, 1992; Good, Maisel, & Kriska, 1998; Wilson, Boni, & Hogg, 1997), memory as a specific contributor to such performance is rarely examined. Police present at a crime scene can be asked to report elements of the scene in their official reports and in courtroom testimony. Such testimony is sometimes required long after the events occurred. We examined the memories of experienced police officers for a simulation of a potentially stressful event--an on-duty shooting incident. Compared to most experimental research, the ecological validity of the present study might increase generalizability of the results to police officers' on-the-job performance, because it involved trained, veteran officers from a law enforcement organization reporting memories of a simulation that was highly relevant to their jobs.
Eyewitness memory research is relevant to this study, but most of it has been conducted with non-work and non-police samples, such as children or college students (e.g. studies by Brady, Poole, Warren, & Jones, 1999; Finger & Pezdek, 1999; Olsson & Justin, 1999; Wells, 1982; see reviews and discussion by Bruck & Ceci, 1999; Fruzzetti, Toland, Teller, & Loftus, 1993; Poole & Lamb, 1998). It can be argued, however, that a professionally trained police officer remembering crime scenes is a special case. It is part of officers' duties to remember and report such events. Compared to most other people, they are often better trained and practised at such memory tasks, and this was true of the officers in the police organization studied here. A recent study showed that trained professionals, including law enforcement officers, can be quite effective at catching deception or lying, although the lay public is not (Ekman, O'Sullivan, & Frank, 1999). Similarly, law enforcement officers have special concern for and training regarding crime scenes. The extent to which eyewitness research on memory conducted with the lay public applies to police officers is unknown, making this an important study both for understanding cognitive work behaviour and more general eyewitness memory research.
Eyewitness research conducted in experimental laboratories has the advantage of strong controls, but the external validity of the results is less certain. The present study was a quasi-experiment, a simulation trading some control in exchange for some ability to simulate crime situations with experienced police officers. Thus, it was higher on ecological validity compared to many laboratory experiments, incorporating parts of a realistic environment into the simulation. In addition, much of the adult eyewitness laboratory research examines memories of bystanders and not people involved in the memory episode. Somewhat more active participation and involvement, a feature of many real-life eyewitness situations, including criminal justice settings, was incorporated into part of the present study.
Four main issues regarding the work-related memory of police officers were addressed: (1) the effects of stressful occupational situations on memory, (2) the effects of review or rehearsal on memory, (3) the relationship between confidence and memory, and (4) the relationship between threat focus and memory. Each of these issues is very relevant to actual police work.
The effect of stress on memory is a controversial subject, and the research results are complex (Christianson, 1992). It is argued that stress might lead to poorer memory; many experts on eyewitness testimony have believed this (survey by Kassin, Ellsworth, & Smith, 1989). Perhaps the extreme examples in this vein include case studies of recovered memories (e.g. Dalenberg, 1997; Schooler, Ambadar, & Bendikson, 1997). On the other hand, research indicates more often that emotional stress is related to better memories, perhaps because of a heightened awareness (e.g. Burke, Heuer, & Reisberg, 1992; Libkuman, Nichols-Whitehead, Griffith, & Thomas, 1999). These conflicting themes give little guidance for conclusions regarding the effects of stress on memories in real organizations.
Chronic stress has been the focus of most occupational stress studies, but Anshel et al. (1997) argued for the need to study acute stressors more in police settings. In one study of acute stress, a simulation study of police training recruits in London (Yuille, Davies, Gibling, Marsen, & Porter, 1994), the accuracy of memories was improved by the mildly stressful event of a suspicious citizen (who turned out to be innocent) being uncooperative and even belligerent. One reason that was an important study is that it examined simulated real-world memory in a sample of trainees rather than of participants who were less relevant to the particular work-related memory situation (such as college students). It is hoped that this should increase the generalizability to real-life situations. To increase ecological validity of the present study, we examined the memories of already trained, experienced police officers from a single organization regarding a work-relevant situation. Much popular literature about stressful events, especially in police work, refers to events that are potentially more traumatic than the one in Yuille et al. (1994) and their effects on memory could be quite different. The Anshel et al. (1997) study of acute stress reported that arresting a violent person, facing an unpredictable situation, facing possible injury, and confronting a person with a weapon were four of the most important acute stressors faced by police officers. The present study simulated this type of stressful event and looked for effects of the event on memories for three elements of a simulated crime: armed people, unarmed people and objects. The armed people were the most task-relevant, the unarmed people the second most task-relevant, and the inanimate objects the least task-relevant for the officers. In this way we could determine the extent to which memories for different elements of a stressful event are more or less accurate. Consistent with most previous experimental research, we hypothesized that (H1) stress will lead to more accurate memory overall.
In memory literature, rehearsal means that someone has previously worked on retrieving the relevant memories (e.g. Pashler & Carrier, 1996). This is often accomplished by testing the participants' memories at an earlier time, before the final memory test. The study by Yuille et al. (1994) is a good example for present purposes because of its focus on police memory. In that study, police recruits were prompted to rehearse by answering questions one week after the incident. In the present study, we administered rehearsal only minutes after the incident, in order to try to simulate the recommended (Grossman & Siddle, 1998) police procedure of taking notes and/or reviewing as soon as possible after an incident. In experimental laboratories, rehearsal consistently leads to better long-term memory, and therefore we hypothesized that (H2) rehearsal would lead to more accurate memory in this situation also.
Confidence in one's memory is important in police work, because others are better convinced of the accuracy of one's memory if confidence is expressed. This can affect juries, e.g. (Chandler & Fisher, 1996; Cutler & Penrod, 1995; Cutler, Penrod, & Dexter, 1989; Granhag, 1997; Lindsay, Read, & Sharma, 1998; Wells, Ferguson, & Lindsay, 1981). Unfortunately, laboratory research has often indicated that confidence is not strongly related to memory accuracy (Bothwell, Deffenbacher, & Brigham, 1987), and this may be especially true when witnessing conditions are controlled to be constant, as in the present study (Lindsay et al., 1998). In their meta-analysis of nearly 35 staged-event studies of confidence-memory relationship, Bothwell et al. (1987) estimated the average effect size at r=.25. Sporer, Penrod, Read, and Cutler (1995) obtained a similar estimate of the overall population effect size of the confidence-accuracy relation (r=.29). The meta-analysis by Sporer at al. (1995) included 30 experimental studies that measured participants' recognition of strangers in a live, staged, filmed or videotaped event.
Because previous studies used non-police participants, it is unknown whether or not police officers' confidence and memory accuracy are related. Training and practice in police work could alter this effect for veteran officers. Good memory is important in some realistic settings such as police work, and officers' experience may cause them to understand the relative accuracy of their own memories better, even though confidence is not clearly related to memory accuracy in the laboratory. The extent to which this might be true is uncertain, but the present study provided evidence on the issue. Based on the previous research, we hypothesized (H3) a small positive or no relationship between confidence and memory for police officers in the present study.
Any stressful event such as shooting involves a threat to the participants. Officers in this study faced an...
|
