Page 2 of 4
The Attention Study: A Study on the Presence
of Selective Attention in Firearms Officers
The study was possible due to the kind contributions of time, energy, and skill from a number of very gracious individuals. These included Dave Blocksidge, Paul Monk, Mark Williams, Dave Jeffries, Patricia Thiem, Dr. Alexis Artwohl, and several staff from the training facility who were kind enough to fill in as “waiting patients” on the benches. Members from the Department of Professional Standards served as interviewers and, of course, armed officers from the Uniformed Occupational Command Units were the volunteer constables who were involved in the study.
The 46 subjects/participants were all armed officers from one of the four armed units within the department: (1) CO6 – The Diplomat Protection Group, (2) CO19 – The Armed Response Unit and Tactical Support Team, (3) CO18 – The Aviation Security Unit, and (4) the Territorial Support Group. Although they varied in age and experience, for the most part they were male and young. Three of the participants were women.
The constables were divided into two reporting groups. One of the groups handwrote reports detailing the incident and their participation in it. The other group was interviewed about the incident using a cognitive interview format.
All of the interviewers were volunteers from the Department of Professional Standards with the Metropolitan Police Service and had training in cognitive interviewing. They were skilled interviewers. Dr. Amina Memon from The University of Aberdeen in Scotland conducted a refresher training session in cognitive interviewing for the interviewers just prior to the start of the research. The interviewers met and individually interviewed each participant. All of the interviews were recorded on audiotape, and these were later transcribed into a typed format.
The other main fashion in which the constables recorded their memory was in a written format. The constables were given incident report booklets, and they recorded their memories of the incident. These were also transcribed into a typed format.
Confer or Nonconfer
Each of the groups—the report group or the interviewed group—were further broken into two groups: (1) confer and (2) nonconfer. The confer groups met immediately after the incident and discussed their roles and perceptions in the scenario. The time for the confer session varied as each group was allowed up to 20 minutes to process the scenario. This method of conferring was not identical to the conferring method currently employed by the Metropolitan Police Service that occurs when their officers are writing their reports. Subsequently, only limited comparisons of the results of this study can be made to the conferring process currently employed.
Mr. Justin Dixon, the Physical Training Manager for the Metropolitan Police Service, arranged to monitor the heart rate of each of the constables by having them wear a Polar chest belt and recorder throughout the scenario and during their reporting of the event in either written reports or interviews. After the event, the constables who were still wearing their heart monitors were given the Metropolitan Police Fitness “Bleep” Test that provided information on their maximum level of physical exertion.
Three cameras stationed at three different locations in the reception area filmed each scenario and each constable. The film from each camera and for each scenario was loaded into a Final Cut Pro Video Editing program that permitted the precise syncing of all three cameras to specific frames. In this fashion, all three videos from each scenario could be viewed simultaneously in a precise frame-by-frame mode.
Transcribed Interviews and Reports
All of the interviews and reports of the constables were transcribed either from the audiotape or from the written report, and each constable’s transcriptions were printed and placed into binders. Each grader was provided with their own binder of the constables’ transcriptions.
The graders were graduates of or graduate students at Minnesota State University, Mankato.
Once the interviews and reports were transcribed, the graders then read the transcriptions of the interviews or reports of the constables and categorized every reported detail into one of the quadrants of attention. This included every item of clothing noted, every person, thought, action, and so on.
After this was completed, the graders individually viewed the synchronized videotapes of each scenario. During this viewing, they evaluated whether the constable was actually capable of reporting each item or behavior on which he or she had reported, and they noted the accuracy of the constable’s report on that item or action. They then recorded the results. The prime area of interest for this project was the focus of each constable’s attention during the high-stress component of this incident. Therefore, all of the analysis on the constables’ attention was taken from the data provided by the constable from the point of the entrance of the hostage taker into the room and the unexpected discharge of the shotgun. The analysis ended when the constables noted that they exited the scenario.
The results from the transcriptions of the constables’ interviews or reports were then coded by the graders and entered into the appropriate categories in Nideffer’s Quadrants of Attention (Nideffer & Sharpe, 1978). The results were grouped by the graders and also by whether the constables were interviewed or wrote a report and whether they conferred or did not confer.
This remembered information then allowed the researchers to extrapolate about the constable’s focus of attention during the incident. If the constable was able to report on something that was present or occurred during the encounter, then it likely was something the constable had observed or focused on during the incident. Inversely phrased, if the constable could not report on something or erroneously reported on it, then it was likely not a focus of the constable’s attention.
The information on which the constables reported was then grouped into one of Nideffer’s Quadrants of Attention, which allowed the researchers, among other things, to determine if there was any applicability of this concept to assist in understanding the constables’ perceptions and memories in a high-stress encounter of limited duration.
Only one constable reported that he had acquired information during the confer stage of the experiment. The reports of all the other constables included details on which they were appropriately focused and on which they were thus capable of reporting. Of significant note here is that constables who conferred, whether they wrote reports or not, even if they did not report learning anything during the conferring, made significantly fewer errors on the material about which they were reporting.
Results from the Cardiac Monitors
The constables’ pulse rates during the incident were monitored by means of Polar cardiac monitors. Mr. Justin Dixon of the Metropolitan Police Physical Education Branch placed the monitors on the constables before they entered the staging room. After the constables completed the scenario, including being interviewed or writing a report, they were tested on the Metropolitan Police Fitness “Bleep” Test to assess their heart rate at maximum effort. Mr. Dixon then collected and analyzed the data. His report on 43 of the constables from this study is included in here as Addendum 1.
The scenario was constructed to be rapidly unfolding, highly dynamic, complex, and very stressful. The cardiac measurements revealed that the researchers had accomplished this goal. The constables’ pulse rates spiked to an average of 75% of their maximum during the conflict situation in the scenario. Very significantly, it also spiked at 65 to 70% of their maximum when they were reporting on the conflict moment during their interviews. For illustration, a constable coded 001 in Group 1 had a maximum pulse rate of 198 beats per minute on the fitness test. During the shotgun blast, his pulse spiked at 159 beats per minute.
Nideffer’s Quadrants of Attention
The research project was primarily designed to assess whether constables in the middle of a violent, rapidly unfolding conflict would develop a perceptual distortion known as tunnel vision. Also investigated was the relevance of applying Nideffer’s Quadrants of Attention philosophy (Nideffer & Sharpe, 1978) to reports from the constables involved in this study, with the goal of assisting researchers in understanding and quantifying a constable’s perception and attention in a stressful, use-of-force encounter in law enforcement. Further, a pilot study was conducted on various means to obtain the memory of a constable after an incident.
The information about the incident provided by each constable, whether the constable was interviewed or wrote reports, was transcribed and then three independent graders graded every word in the transcription. Every item of clothing, thought, behavior, movement, etc., in the scenario, as noted by each constable, was graded for its appropriate fit into one of Nideffer’s Quadrants of Attention. For example, if the constable noted the double barrels of the shotgun, it was graded as being an item that was entered into the External Narrow focus of attention. If the constable thought they were in danger, it was an Internal Broad focus. If the constable thought he had to move to the pillar, it was an Internal Narrow focus. If the constable noted the assailant fell to the floor, it was an External Broad focus. The data was then entered into the appropriate attentional quadrant. This then gave the researchers a numerical indicator of the number of items the constables accurately remembered when they were engaged in the simulated critical incident and the placement of those items in Nideffer’s Quadrants of Attention. It is obvious that the constables could not see everything occurring at any instant in this encounter, but the items that the constables remembered at the end of the incident are a good indicator of their focus of attention during the incident.
A variety of statistical analyses were then applied to the data. However, in the author’s very first undergraduate class in research, the students were encouraged to “eyeball” the data first to see if any trends in the data “popped out” at them. The following tables are presented in that spirit.
The purpose of presenting the data in this format is to illustrate the blatantly obvious nature of this data. The above table illustrates the average score in each quadrant for the average constable when the data is combined across all of the research conditions. For example, the reader will note that the External Narrow quadrant of attention generally has nine times more data recorded in it than the Internal Broad quadrant. The Internal Narrow quadrant is generally over twice as large as the Internal Broad quadrant. The External Narrow quadrant is roughly about twice as large as the External Broad quadrant. Therefore, the researchers can state unequivocally that during the most stressful component of this simulation, the constables experienced a narrow focus of attention and were externally tunneled specifically on items or behavior that were crucial for their performance in this situation. This tunnel vision or selective attention is characteristic of all humans who engage in this type of encounter, whether they have received law enforcement training or not. This study illustrates that this phenomenon also occurs in well-trained peace officers.
The following is presented for a more detailed analysis of the data. Tables and figures reflecting a higher level of analysis are included in the Addenda and are available for review.
Condition I: Nonconferred and Wrote Reports (11 Constables)
A total of 11 constables were in this group. Immediately after participating in the scenario, each of the teams of constables in this experimental condition returned to the briefing room, where they returned their equipment. They were immediately separated after this. They were instructed to not discuss the incident with anyone. They were then told to go to an area where they could write their reports and that they should turn them in when they were completed. At some point after writing their reports and during the same shift, they took their fitness test.
The following table includes the average scores for the constables, recorded in each of the four quadrants of attention for this group. The numbers listed in each quadrant of the chart are averages in that quadrant for all of the scores of the 11 constables across all three graders.
The constables who wrote reports in this category provided the least amount of information despite being requested to write full and complete reports about everything in the incident. They reported on the essence of the incident, providing little extra information. Their descriptions of the incident and the subjects, even when they were mentioned, were the barest, with little or no elaboration on such things as behavior or clothing. The constables’ reports on their thought processes during the incident were also extremely sparse. Note that the average score per constable was fewer than two thoughts per constable per incident for an Internal Broad focus.
The reader should keep in mind that the pattern of the constables’ scores regardless of the number of items listed is overwhelmingly large in the External category and exceptionally larger by far in the External Narrow category as compared to the External Broad category.
Simply stated, the constables who wrote reports primarily reported on items that they were externally and narrowly focused on while they were engaged in attempting to respond and control the events in the scenario.
Fourteen constables were in this group. The teams of constables in this experimental condition also immediately returned to the briefing room where they returned their equipment. They were then allowed to discuss the incident for a brief period of time (no more than 20 minutes) and then were instructed to go to an area where they could write their reports and turn their reports in when they were completed. They were also instructed that after the initial conferring, they were not to discuss the incident further with anyone. At some point after writing their report and during the same shift, they took their fitness test.
Constables in this experimental condition, on average, reported the same amount of data in all categories as the constables who wrote reports and did not confer.
Comparison Between Conferring and Nonconferring in the Report Category
Generally, the researchers would not have expected to have a difference in the Internal category between these two groups as this category relates to the constables’ internal reflections and awareness of their own behavior. Considering that most of the constables did not mention their thoughts or make comments about their own behavior during the conferring and that other constables did not elicit their thoughts, there should be no difference in this category between Condition I and Condition II. If there was any expected increase, it should have occurred in the External category because of the conferring. However, during the conferring process, the researchers observed that the constables did not appear to learn anything “new” during the conferring. Therefore, it is not surprising that there was not a difference between conferring and nonconferring in any of the quadrants in the Report category. This is particularly true when what the constables reported was so sparse.
The most significant difference between the conferring and nonconferring report groups occurred in the number of errors that each group made. This is explained more completely in the “Error Rates” section. It is important to note here that in the External Narrow quadrant (which is the quadrant that the constables would be noting the behavior of the subject, the threat presented, etc.), the group that did not confer had four times more errors in their written reports than the group that conferred. This may mean that even though the constables did not comment that they learned anything new during the conferring, some accurate information processing was occurring.
Condition III: Nonconferred and Interviewed (6 Constables)
Six constables were in this group. The constables in this group returned to the briefing room right after the scenario and turned in their equipment. They were immediately separated after this and asked to not discuss the incident with anyone. They were then instructed to go to an area where they could be individually interviewed using a cognitive interview format. At some point after being interviewed and during the same shift, they took their fitness test.
Constables in both of the interviewing conditions provided much more information overall than did constables who wrote reports. This was particularly true in the Internal quadrant. The added information was generally two to four times the amount of information provided by the report group, dependent upon the quadrant. In the External quadrant, the interviewed constables generally provided about twice the information as the constables who wrote reports. The reader will note, however, that particularly in the External Narrow quadrant, which was the primary focus area of all of the constables in the scenario, the error rates of those constables who were interviewed was exceptionally high in comparison to those who wrote reports (see the section on “Error Rates” for more information).
Note that the pattern of the constables’ scores regardless of the number of items listed in each quadrant is overwhelmingly large in the External category versus the Internal category and exceptionally higher by far in the External Narrow as compared to the External Broad quadrant. This means that regardless of how the constables reported their information, all the constables had experienced the same type of perceptual process.
Condition IV: Conferred and Interviewed (15 Constables)
Fifteen constables were in this group. The constables in this experimental condition also returned to the briefing room where they returned their equipment. They then conferred with each other immediately after the incident, discussing the elements of the incident for a brief period of time. Sometimes the discussions lasted up to 15 or 20 minutes. They did not write a report. After conferring, they were individually interviewed using a cognitive interview format. At some point after being interviewed, and during the same shift, they took their fitness test.