Photo: UTM Media Dept, 2016 Editor Brittani Schroeder spoke with Judith Andersen, an Associate Professor at the University of Toronto Mississauga, about her studies of the psychophysiology of performance and health among individuals in high-risk occupations.
Q: How did you come to research policing and use-of-force?
I maintain a research program as the Director of the Health, Adaptation, Research on Trauma (HART) lab. For the past decade, I’ve teamed up with academics and police agencies in North America and Europe to conduct a variety of studies—for example, randomized controlled trials and those with long-term follow up.
What we quickly realized during these studies is that stress not only impacts health over time, but it really can influence the application of skills, including de-escalation, the use-of-force and lethal force decision-making during critical incidents. Of course, these findings have strong implications for police and public safety. As well, my team and I have had the honour of working on provincial and federal policy on use-of-force and de-escalation.
Q: Can you tell us about your recent study?
Our study examined police lethal force errors and stress physiology during simulated evaluation scenarios that were presented in video and live formats. In a sample of 187 police officers, we examined the rate of use-of-force errors during video and live scenarios, and we measured their heart rates in real time as a reflection of stress response activation. Although there were low error rates overall, it’s important to note that any errors made during an evaluation may have direct professional consequences for the officer (e.g., loss of authorities). Thus, it’s very important to make sure that error rates are not caused by the design or delivery of the scenarios, but rather reflect the officer’s skills.
It’s important to note that these simulations reflected realistic calls for service in a large, representative police agency in Canada. The research was an observational field study of current police practices, not a lab-based experiment. That is an important distinction – while the live and video scenarios were not identical in their content, using observational methods is an important and valid approach to learning about behaviour in the environment in which it occurs. This method enhances ‘ecological validity’, increasing the confidence that the conclusions of the study are relevant to police scenario-based training and evaluation.
Q: What were the results of the study?
We noticed that there were significantly fewer lethal use-of-force errors in live versus video scenarios. While there may be multiple factors that impact errors, we believe the data indicates two likely reasons:
- The officers were more engaged during the live scenarios, as evidenced by greater physiological reactivity—change in Heart Rate (HR) from rest—compared to the video scenarios.
- Neuroscientific research indicates that two-dimensional video encounters, similar to the video scenarios used by the police agency in this study, are processed by different brain networks than live interactions, which may impact the person’s sensory awareness and subsequent responses to a virtual versus live encounter.
Let me expand upon each of these in turn.
For the first factor, when a person faces a challenging situation with potential threat (as is the case in police encounters), the brain sends signals to the body to prepare to meet the demands of the situation, whatever they may be. This means that the stress response system is activated. Not all activation is bad – in fact, a moderate amount of activation increases cognitive and sensory awareness, attention and vigilance, and is advantageous to meeting the demands of the situation.
Medical science research on real-time HR, in combination with direct observation of a person’s actions, show that about a 10-point increase in HR in the absence of significant physical movement indicates psychological and physiological engagement of the stress response system. While overactivation of this system (known as ‘flight or flight’) is helpful when survival is on the line, it can interfere with social and verbal skills when activation is excessive or prolonged.
In both the video and live scenario conditions, we observed a significant increase in stress response engagement; not overactivation, but instead, a sign that officers were paying attention and in a state of ready to perform their skills. However, we did observe a higher amount of reactivity in the live condition that indicates even more engagement compared to the officers’ own HR in the video condition. It may be the case that the live scenarios better reflected the full sensory context in which an officer applies their skills in the real world, and therefore raised their physiological activation in preparation to meet the demands of the live simulation.
This notion is supported by the second factor I raised.
Neuroscientific research shows that subtle facial muscle expressions, movements and speech patterns are processed more effectively during live versus virtual interactions; you can reference Redcay et al., 2010, and Tomasello et al., 2005. The human brain is finely tuned to detect visual information like direction and speed. When this information doesn’t match with other senses like sound, or doesn’t match the direction and speed of an officer’s own movement, virtual simulation environments may feel ‘unnatural’ or even induce negative effects like cybersickness; look at Giessing, 2021, and Kleygrewe et al., 2023. This lack of realism may account for the relatively higher errors observed in our study, as officers may have been more cautious in their behaviours towards a ‘real’ person standing before them.
Anecdotally, officers have mentioned to our research team that they do not prefer video simulations when they are more basic (e.g., video projections on a wall versus immersive virtual (VR) or augmented reality (AR) systems). Officers noted that they would rather do live scenarios in order to have the whole context be more real and use all of their senses and skills as they would in the field. Although officer-reported preferences were not a part of this study, they are an important aspect to consider in future research.
Q: What can police agencies take away from this study?
Many virtual simulation configurations used by police agencies (including the one used in the current study) are relatively ‘basic’ compared to advanced VR and AR systems and may influence performance outcomes by presenting physical and perceptual limitations that do not exist in live simulations or real-world interactions. Therefore, agencies must be mindful of whether learning or evaluation goals could be influenced by the way scenarios are delivered (either for training or assessment purposes), especially if virtual equipment (e.g., Taser, firearm) is incompatible with officers’ duty equipment.
Top of the line VR requires financial investment, technical expertise and time resources to keep it running effectively, including continuous updates to software and hardware. These are barriers for some services in Canada and can limit accessibility to the use of more realistic virtual training and evaluation platforms. The good news is that live scenarios delivered in more simplistic settings (e.g., agency training rooms, facilities, empty buildings in the community) do test a full range of skills, including de-escalation and use-of-force decision-making.
In line with international police researchers, my team and I would recommend that scenarios are delivered in a format (either virtual or live) that: a) most accurately represents the context in which the skill or competency of interest will be demonstrated in the field, and b) leverages agency resources to be most efficient. Prioritizing the latter does not prioritize officer learning, evaluation or operational safety.
Print this page
