**Protocol** ------------------------------------------------------------------------ **The effects of tension level during a traffic stop on noticing of a conspicuous threat (a gun)** ---------- General Information ------------------- Participants will be adult trainees at the Illinois Police Training Institute located in Savoy, IL. A new group of trainees attend the institute every 12 weeks for intensive training in police procedures, best practices, ethics, self-defense, etc. A core part of this training involves participating in simulated interactions with civilians under naturalistic and realistic conditions. Officers are taught how to conduct a traffic stop, how to defuse a tense situation, how to respond to threats, etc. Over the course of the training, they perform a number of simulated traffic stops in which an experienced actor plays the role of the driver. Trainees are informed about the type of stop they are making (e.g., suspect in a crime, routine traffic stop such as failing to stop at a stop sign, etc.). The training scenarios vary in intensity, but they are designed to teach defensive tactics, appropriate scanning methods, and interpersonal techniques that can help defuse a threat. The actor in our study is an experienced actor who has been role playing in these training simulations for years. He is practiced at simulating a compliant driver as well as an aggressive/hostile one. Our scenario will be a routine, solo traffic stop. (In most of the training scenarios, other officers observe as one officer performs, but in this case, only the trainer will observe). Subjects will be told that they have just pulled over a driver for failing to stop at a stop sign and that they will be issuing a ticket. They are taught that anything might happen during a stop and that they should always use best practices to protect their own safety and that of others. The subject will approach the car and initiate an interaction, and the driver will either respond in a congenial way or aggressively. In both cases, the driver will follow instructions (e.g., providing their license and registration), but in the aggressive case, they will be verbally abusive and overtly upset. A gun will be positioned on top of the dashboard in front of the passenger seat. The core question is whether or not the subject notices the gun. Procedure --------------------------------- Trainees will be tested individually. The testing will take place several weeks into the training program, after the trainees have had experience with multiple traffic stops, but before they have engaged in high-risk apprehensions. They are familiar with standard procedures for approaching a driver, obtaining the driver's license and registration, examining the car for possible threats or illicit objects, dealing with a compliant or aggressive driver, etc. They have practiced such stops and are taught how to handle themselves and their interaction. In the files section, we will add video of Mike Schlosser approaching the driver and conducting a standard traffic stop and ticketing procedure. He is the primary trainer for the protocol for such traffic stops. Trainees participating in the study will be excused from a classroom setting and asked to go to a squad car positioned behind a "stopped" car. They will be recruited by Mike Schlosser or by a previous subject (who returns to the class and sends someone else out -- the trainees are used to doing this as part of their regular training). Critically, the "recruiter" will not know the condition for that participant. The trainee will be told: > "You have just pulled over this car because the driver failed to stop > at a stop sign. You will be issuing a ticket. Go ahead and go > through the process." The exact wording might vary slightly so that it sounds natural, but the script above conveys the content. Mike Schlosser will not know whether the driver will be aggressive or compliant at the time he gives the instructions. The driver in the car will have a printed list with the condition for each participant (compliant/aggressive) in a randomly generated sequence. A gun will be positioned on top of the dashboard in front of the passenger seat such that it is visible through the driver's window from the officer's perspective. Video and photos illustrating the gun's position and visibility will be added to the Files part of this project. The trainee will go through the procedure of approaching the vehicle and interacting with the driver, requesting their license and registration. If they do not notice the gun, they will return to their car to issue a ticket. If they do notice the gun, they can take a number of options. They can retreat quickly to the shelter of their own car and issue orders from there. They can draw their own gun and order the driver out of the car. They can calmly try to defuse the situation by getting the driver to exit the car. Once the officer takes any of these actions, Mike Schlosser will end the scenario and will send the officer to the experimenter who had been positioned out of sight of the cars and Mike. The experimenters will then obtain consent to ask questions about the traffic stop. The experimenter will say: > "In addition to giving you practice with a traffic stop, the scenario > you just completed is also part of a study we are conducting of how officers perceive, think about, and remember > traffic stops. Would you be willing to answer some questions for us?" If they say yes, they will be given a consent form to read and sign and the experimenter will answer any questions. The experimenter then will read the following questions in sequence and the experimenter will record their answers on a response sheet. After any "Yes" answer, the experimenter will ask for elaboration if the subject doesn't provide it spontaneously. For example, if the subject notes that there was something potentially dangerous but doesn't mention a gun or something else of danger (e.g., tinted windows), the experimenter will ask them "what did you find potentially dangerous" and will follow up as necessary. The experimenter will write all answers on the response sheet. 1. During the traffic stop, did you notice anything that might have been a danger to you? 2. Did you notice any weapons? [NOTE - if the subject reported the gun in response to question 1, the experimenter will skip this question] 3. Did you notice any drug paraphernalia [Note that the car contained no drug paraphernalia. The drug question is intended to measure whether trainees might respond "yes" even if they did not actually see anything illicit.] 4. Was the driver aggressive in any way? 5. In a real traffic stop, would you have felt in danger from a driver like the one you stopped? 6. Have you heard about this particular traffic stop scenario from other trainees? Data from any participants who knew in advance that a gun would be present will be excluded from the analyses. The experimenter will then ask for the subject's age and whether or not they have any patrol experience, noting both on the response sheet. Once they have answered these questions, the experimenter will explain the purpose of the study and our hypotheses and will describe the nature of inattentional blindness and why better understanding what people notice could potentially help with future training. For those who did not notice the gun, we will inform them of its presence and (at their option) show them where it was in the car. We will also inform them that there were no drug paraphernalia in the car. Finally, the experimenter will emphasize the importance of not discussing the scenario or the study with any of the other trainees or with their home departments and will tell them that, if asked, they should just describe what they did as a solo traffic stop scenario. Debriefing will be spoken rather than written, and all questions will be answered. Participants and stoping rule --------------------------------- Each 12-week training program includes 50-100 trainees. We will have either a half day (4 hours) or a full day (8 hours) with each group of trainees in a program. Based on piloting, we can test approximately 6-8 subjects per hour. Consequently testing will take place across multiple groups of trainees. As a result, some testing take place in late fall and other testing will take place in the winter, spring, or summer. Based on the reported pilot testing, we expect noticing rates of 25-50%. We will test a minimum of 80 participants with at least 40 in the "aggressive" condition and 40 in the "compliant" condition. We will continue testing new recruits during each new training program session until we meet the minimum of 40 in each condition. Once we have reached our minimum sample size, we will continue testing that group of trainees until the end of that testing session and then will cease data collection. With this approach, we could have a final sample with as many as 180 participants, although we do not anticipate testing more than about 100 participants. Approximately 70-80% of the trainees are male, so we expect roughly that distribution in our final sample. To randomize the participants, we first generated a list of 80 condition labels (40 in each condition). We then randomized that list in Excel by pasting one per row, generating a column of random numbers next it, and then sorting by the random numbers. We will then assign participants to conditions in that order. Given that we likely will need to test more than 80 participants, we repeated this process for the next 80 subjects. We likely would not use that full second set of condition orders, so it's possible that we will end up with fewer subjects in one condition than the other. **The ordering for subjects 1-80 is:** Compliant, Compliant, Aggressive, Compliant, Aggressive, Aggressive, Aggressive, Compliant, Compliant, Compliant, Aggressive, Compliant, Aggressive, Compliant, Aggressive, Aggressive, Compliant, Compliant, Compliant, Aggressive, Compliant, Compliant, Compliant, Compliant, Compliant, Aggressive, Compliant, Aggressive, Aggressive, Compliant, Aggressive, Aggressive, Aggressive, Aggressive, Compliant, Aggressive, Compliant, Aggressive, Compliant, Aggressive, Aggressive, Compliant, Aggressive, Aggressive, Aggressive, Compliant, Compliant, Aggressive, Aggressive, Aggressive, Aggressive, Compliant, Aggressive, Compliant, Aggressive, Aggressive, Aggressive, Aggressive, Aggressive, Aggressive, Compliant, Aggressive, Compliant, Compliant, Aggressive, Compliant, Compliant, Compliant, Compliant, Compliant, Aggressive, Compliant, Compliant, Compliant, Aggressive, Compliant, Aggressive, Compliant, Aggressive, Compliant **The ordering for 81-160:** Compliant, Compliant, Compliant, Compliant, Compliant, Compliant, Aggressive, Aggressive, Aggressive, Aggressive, Aggressive, Compliant, Aggressive, Compliant, Compliant, Aggressive, Aggressive, Compliant, Compliant, Aggressive, Aggressive, Compliant, Compliant, Compliant, Compliant, Aggressive, Compliant, Compliant, Aggressive, Aggressive, Compliant, Aggressive, Aggressive, Aggressive, Aggressive, Compliant, Compliant, Aggressive, Compliant, Aggressive, Aggressive, Aggressive, Compliant, Compliant, Aggressive, Aggressive, Aggressive, Compliant, Compliant, Aggressive, Aggressive, Compliant, Aggressive, Aggressive, Compliant, Compliant, Compliant, Compliant, Aggressive, Aggressive, Compliant, Aggressive, Aggressive, Compliant, Aggressive, Compliant, Aggressive, Compliant, Aggressive, Compliant, Aggressive, Compliant, Aggressive, Compliant, Aggressive, Compliant, Aggressive, Compliant, Compliant, Aggressive, If it looks like we will need more than 160 participants, we will generate the additional condition assignments in the same manner. Data Inclusion ------------------ Although all trainees recruited for the study will participate in the traffic stop scenario (as part of their training), they are not required to provide data for our study. Immediately after the end of their traffic stop, trainees will be directed to the experimenter where they will be informed about the study and asked if they are willing to answer our questions. If they assent, they will be given a consent form to read and sign and the experimenter will answer their questions. Only those who consent will be asked further questions, but all will be asked not to discuss what they experienced and will be told the core hypotheses of our study. Data Exclusion ------------------ For those who agreed to participate and answered our questions, data from any trainees who admit to having heard about the study in advance will be excluded from all analyses and replaced with data from additional trainees. We also exclude participants due to experimenter error (e.g., failure to put the gun in place). That decision will be made before tallying any results, and the actor will note any problems with that subject on the condition order sheet. Coding and Analyses ------------------ The dependent measure in this study is the percentage of people in each condition who noticed the gun. Given the nature of the unexpected object and the requirements of police procedure, we do not expect any ambiguity in coding whether or not subjects noticed the gun (and pilot testing bears that out). Anyone who mentions the gun during questioning will be counted as having noticed and anyone who does not report noticing a gun will be counted as having missed it. We will first calculate the total percentage of participants who noticed the gun (ignoring condition). We then will calculate separately the percentage of people who noticed in the "aggressive" condition and in the "compliant" condition. The overall noticing rate does not require any statistical analysis -- the percentage is the critical information. If more than zero miss the gun, that data point provides the effect. For the comparison of noticing in the two groups, we will conduct a two-tailed Z test to compare the proportion of people in each group who noticed the gun: ![Z equation][1] - rho1 is the proportion noticing in condition 1 - rho2 is the proportion noticing in condition 2, - rho in the denominator is the overall proportion noticing - n1 and n2 are the number of subjects in each condition respectively - (note: Formula image from [Wiley][2]) We will then compute the 2-tailed probability for that Z value given that our prediction is two tailed (we would be interested if people in the Aggressive condition noticed more or less than those in the compliant condition. With a sample size of 40 in each condition, a difference in noticing rates of about 25% would be statistically significant. Finally, we will look at sex and age differences in noticing as well as any effect of patrol experience. We have no firm hypotheses for these demographic variables, so these analyses should be treated as preliminary and exploratory. [1]: http://media.wiley.com/Lux/09/360009.image0.png [2]: http://www.dummies.com/how-to/content/how-to-compare-two-population-proportions.html