How far have we advanced in simulation-based training for policing and law enforcement? A literature review 2014–24

Abstract

The twenty-first century demands on police and law enforcement similarly demand effective, efficient and relevant officer training endeavours. Guided by the PRISMA literature review model this study identified and analysed a total of 35 peer-reviewed studies (2014–24) centred on simulation scenario-based training offering insight into (1) what are the different types and platforms used in scenario-based simulation training and (2) what skills are being trained in simulation-based exercises. The findings revealed a preference for high-fidelity technological tools for short training scenarios centred on tactical and procedural knowledge. Conversely, incidents that require conversations and negotiations remain restricted to platforms with limited technological integration or live role-players. The review identified (1) skills for daily operations, decision making, de-escalation training, major incident management were those most commonly attributed to simulation-based training and (2) the extent to which virtual environments can replicate major incidents that are sufficiently immersive for training continues to be a work in progress.

INTRODUCTION

The sustained increase of societal demands on police agencies, similarly, demands agile, relevant, efficient and effect training approaches for police and law enforcement officers. The adoption of simulation-based learning across multiple professions including policing and law enforcement has contributed to supporting placing the learner in authentic, real world replicated exercises and environments in preparation for operational demands. Studies of the application of a simulation-based learning approach for developing specific profession competencies and confidence are spread across the vast array of professional activities. For example, the work of Rogers and Franklin (2021), Sipiyaruk et al. (2023), and Dubé et al. (2020) discuss the application of simulation-based learning for the medical profession. The work of Fussell and Truong (2020) and Hon et al. (2020) discuss the application of simulation-based learning for aviation professionals. Yang et al. (2020) and Harris et al. (2023) discuss the use of simulation-based learning in defence training to name a few. There is a plethora of published work in the simulation-based learning domain (in professions other than policing and law enforcement) and as this approach to learning continues to gain momentum it is anticipated the study of diverse aspects of this learning application will similarly expand.

The policing and law enforcement education and training domain is witnessing an ever-increasing trajectory of applying technologies such as Extended Reality (XR) and Mixed Reality (MR) to meet the demands of developing the preparedness of officers for the challenges of maintaining community safety in their operational duties (Giessing 2021; Caserman et al. 2022; Zechner et al. 2023). The diverse nature of policing and law enforcement and, by association, the training needs in this field are increasing the body of literature devoted to the application of simulation-based learning as indicated in this literature review. In parallel there is not as yet a synthesised study of this body of literature that enables an organised presentation of the different types of platforms used in scenario-based simulation-based training for police and law enforcement and further, which skills are being trained through this approach to learning delivery.

This focus of the literature review is guided by the work of Gray (2019), Davies and Krame (2023), Brydges et al. (2010), and Lefor et al. (2020) for example, who advocate a pivotal factor for effective simulations is the consideration of the requisite levels of fidelity to create realistic behaviours. In addition, with the role of police and law enforcement officers demanding readiness to meet a diverse range of operational challenges both as individual officers and in teams, it is valuable to acknowledge that different types of skills draw on different training simulation tools. A central tenant of simulation-based learning as indicated in the following literature review, is to place the student at the centre of the learning with the facilitator/teacher as a guide on the side and this pedagogical approach enabling the learner to take responsibility for their decisions and actions (Levin 2024).

This literature review aims to contribute to future education and training endeavours for policing and law enforcement, through identifying studies over the past decade that focus on the application of simulation-based learning for police and law enforcement officers. Specifically, the review aims to answer the following questions in the context of policing and law enforcement education and training: (1) What are the different types and platforms used in scenario-based simulation training? and (2) What skills are being trained in simulation-based exercises?

METHODOLOGY

This review adopted the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) model where a descriptive summary of key issues is first identified, before the research is reviewed, categorised and synthesised. Analysis is done both within a study and between studies to highlight how simulation tools are used to train law enforcement officers, and various findings and issues that might have surfaced with the adoption of such tools (Page et al. 2021).

Procedures

A systematic search of literature on the use of simulation for police and law enforcement using the databases from Oxford Academic Press, SAGE journals, Wiley Online Library and Taylor & Francis was conducted. The selection of these databases is based on their propensity to offer police and law enforcement journal offerings. A manual hand search was also done to identify relevant studies that might have been missed through traditional searches. Only peer-reviewed journal articles written in English were reviewed.

Search strategy

The search strings used to locate journal articles with empirical evidence for the use of simulation-based learning technologies for law enforcement training, particularly in the field of incident management, were: (Law enforcement training) OR (police training); AND (simulation training) OR (simulation-based training) OR (virtual) OR (simulation); AND (incident management) OR (critical incident management) OR (ground incident management).

These search strings allow different dimensions when adopting training simulation tools to be identified. Articles that were written between January 2014 and June 2024 were chosen as the period encapsulates the development of how the use of simulation in police training has evolved over the past decade, especially with the growth of new technologies such as Virtual Reality (VR), Augmented Reality (AR) and MR. Table 1 presents the inclusion and exclusion criteria.

Table 1.

Inclusion and exclusion criteria of papers for review.

Inclusion criteria	Exclusion criteria
Articles written in English	Articles not written in English
Articles published 01.2014 – 06.2024	Articles published outside the identified time frame
Peer reviewed journal articles	Books, commentary items, grey literature
Empirical studies involving the use of simulation for training	Concept papers describing how simulation can be used for law enforcement training
Clear description of how simulation is adopted for law enforcement training.	Studies focused on training for medical personnel
Studies that presented data associated with officers’ decision-making processes in incident or crisis management	Studies that do not use any form of simulation for incident or crisis management training
Studies that offer insights on officers’ perceptions of the use of simulation for skill development	Frameworks and design consideration for crisis management training

Guided by the PRISMA quality standards, the results identified 3,325 articles screened after 66 duplicative records were removed from the original 3,391 papers that were extracted. After removing articles based on abstracts and titles, a total of 91 reports were retrieved to assess their eligibility. A final list of 37 articles were selected for review. The extraction process, based on the PRISMA 2020 statement (Page et al. 2021), is presented in Figure 1. The articles and summary of coding are found in Appendix  1.

Figure 1

Screening of articles using the PRISMA 2020 process

DATA ANALYSIS

Descriptive

The average number of articles published in the 10-year period was 3.7 papers per year with the exception of 2014, 2022 and 2024 in which five papers were published that met the inclusion criteria. There was no sharp increase or dip in number of papers published from 2019 to 2022 where there were four papers published each year in 2019, 2020 and 2021 (Figure 2). Of note, this potentially signifies that police training utilising simulation did not come to a standstill during the global COVID-19 pandemic; however, consideration needs to include the potential for delays/lags in publication processes resulting in studies undertaken prior to 2020 being released during the 2020–2021 period.

Figure 2

Number of articles published in each year

There appears to be an uptake in adopting critical incident simulation for police and law enforcement training in 2024 as five studies (Di Nota et al. 2024; Kleygrewe et al. 2024a, 2024b; Stenshol et al. 2024) have been published in the first half of the year. This figure for the first 6 months of 2024 is higher than the average achieved across a full year in the previous years. Additional descriptive statistics that arose from the analysis are presented in Table 2.

Table 2.

Descriptive statistics of analysed papers.

Characteristic	N	% of total articles (N = 37)
Site of study
Europe	18	48
North America	13	35
Australasia (Australia and New Zealand)	4	11
Asia	1	3
Unstated	1	3
Total	37	100
Profile
Police Officers	32	86
Multi Agency	5	14
Total	37	100

Research studies from Europe contributed to approximately 50% (n = 18) of the review, where there are five papers that described how scenario-based simulation is conducted in the United Kingdom, four from Norway and Sweden, two from Switzerland, one study each from France, Netherlands and Italy. The spread of studies involving the use of simulation for police training across Europe also symbolises the readiness of Europeans to share and contribute knowledge to the global police and law enforcement education and training community. Countries in North America contributed 36% (n = 13), where a majority of the studies were based in the United States of America (n = 8), whilst the remaining five studies were based in Canada. All four papers (Davies 2017; Hine et al. 2018, 2019) from Australasia were based on studies done in Australia, whilst the only study done in Asia was conducted in China (Liu et al. 2018).

As the focus of this review aims to look at the use of simulation for police and law enforcement training, the majority of the papers (n = 32, 86%) involve the training of only police officers. However, as many critical incidents involve multi-agency cooperation, five papers (Wilson and Gosiewska 2014; Malizia 2016; Waring 2019; Korn et al. 2021; Mattingsdal et al. 2023) were also reviewed as they involved police departments training with other agencies to de-escalate large-scale civil emergencies. As the military and police are often the ones to lead the management of large-scale civil emergencies, Mattingsdal et al. (2023) included comparing similarities and differences of commanders from both police and military backgrounds in their decision-making process.

Types of simulation tools

In law enforcement training, simulations contribute to the development of professional intuition and decision-making through structured experiences in environments that have been designed to demonstrate officers’ internalisation of training (Phelps et al. 2016). An integral aspect of designing a simulation that reflects reality, whether of live or virtual, must be to mimic the stress that the officer will face in the field (Baldwin et al. 2022). Broadly speaking, there are two types of simulation: live and virtual simulation. Traditionally, live simulation involves role players, where officers, healthcare professionals, and members of the public take on certain roles to simulate the dynamic nature of crises. Over the past decade, the evolvement of technology-enabled simulation has resulted in higher fidelity. Emerging technological developments and platforms such as VR and VirTra systems allow high-fidelity simulations to be created in the virtual space. High-fidelity simulations are critical for skills training and psychological experimentation purposes as they minimise the disjoint between reality and the simulation environment, making them more effective. As the type of simulation adoption depends largely on the purpose of the skill training, this review surfaced four large categories of simulation exercises. The breakdown of each type is appended in Table 3.

Table 3.

Type of simulation.

Nomenclature	N	% of total articles (N = 37)
Live Simulation	13	35
Computer Based Virtual Simulations (n = 9)
Interactive	5	14
Strategic	2	5
Video recording	2	5
Highly Immersive Virtual Simulations (n = 12)
Virtual reality (head mounted devices)	5	14
2D large screen simulators (VirTra, MILO, etc.)	6	16
Virtual Reality and VirTra	1	3
Table-top Exercises	1	3
Hybrid—combination of modalities	2	5
Total	37	100

Live simulation

Despite the advancement of technology, simulation exercises involving role players remain to be the majority (n = 13, 35%) of all the research conducted. Live simulations involve in-person interaction between an officer responding to actors, peers or instructions who are delivering scripted scenarios in staged environments (Di Nota et al. 2024). These studies usually require police to negotiate with perpetrators to de-escalate the situation and may or may not have a ‘correct’ approach to resolve the situation. The uncertainty of human behaviour thus requires live role-players to take on roles. Such scenarios include hostage situations (van den Heuvel et al. 2014), domestic violence cases (Cotard and Michinov 2018; Hine et al. 2018), mental health crisis situations (Alvarez 2020; Helfgott et al. 2020) or the management of intoxicated persons (Paquette and Prince 2021). Live simulation exercises have also been shown to improve officers’ proficiency in critical decision-making and confidence in communication and reduce social stigma amongst police when dealing with individuals in mental health crises (Krameddine et al. 2013). The realism of live simulation exercises appears to be a preferred option for familiarising officers with specific incident management process (Armstrong et al. 2014; Sjöberg 2014; Sjöberg et al. 2015; Johnsen et al. 2017). Live simulations were also found to be more effective in evaluating use of force (UOF) competencies that involve: (1) more physical activity when utilising field equipment; and (2) non-verbal and communication skills, such as interactions with members of the public (Di Nota et al. 2024).

Virtual simulations

Virtual simulations involve a wide range of tools and systems to allow officers to experience a semblance of operational reality during training to prepare them for work. It comprises a wide range of formats, from basic video-based desktop configurations to highly immersive systems such as VR or CAVE systems. Virtual simulations allow for variability in individual scenarios and maximise resource allocation (Zechner et al. 2023). They allow officers to be exposed to scenarios with repetition and are optimal for drills of discrete, component skills that need to be performed in mentally and physiologically demanding contexts (Di Nota et al. 2024). Nine research studies (25%) explored basic desktop-based virtual configurations, which are often deployed as they are less costly but serve their purpose of putting officers through simulated exercises. These simulations can be interactive (Biggs et al. 2015; Söderström et al. 2019; Scott et al. 2022; Mattingsdal et al. 2023) or encourage officers to reflect (Phelps et al. 2016; Söderström et al. 2022). Conventional table-top exercises that involve 3D mock-ups, props and materials can also be translated into computer-based virtual exercises (Wilson and Gosiewska 2014; Korn et al. 2021), where injects can be given in various forms, such as simulated reports, media posts and video recordings. As these simulations are virtually executed, different injects and pathways towards different outcomes can be created for officers to experience the psychological stress of managing an evolving crisis.

Highly immersive and high-fidelity simulation systems have been known to be able to recreate real-life dangerous situations by providing multi-sensory experiences through audio-visual stimuli. Three papers (Davies 2017; Cook et al. 2022; Kleygrewe et al. 2023) adopted VirTra systems to understand officers’ perceptions and responses within highly immersive 2D environment. The VirTra system has five large screens arranged in 300 degrees and participants stand within the space in front of the screen with the ability to move freely during the scenario. Screen images were static, and the participants’ views did not change with their movement. However, it stores a range of pre-recorded videos containing different branching options. This gives instructors the flexibility to choose how the scenario progresses depending on the decisions made by officers. VirTra systems allow external trackers and devices to be connected to provide haptic feedback during the exercise. According to the work of Adilkhanov et al. (2022), the word haptic(s)

…refers to the capability to sense a natural or synthetic mechanical environment through touch. The last decade has seen a dramatic increase of haptic devices, driven by application domains such as haptic robot teleoperation, virtual reality (VR) and augmented reality (AR).

In Kleygrewe et al. (2023) study, an electric feedback device was attached to the participant and triggers were delivered by the trainer to simulate pain stimulus when the participant was shot by a perpetrator. Whilst in the Cook et al. (2022) study, a modified Glock 17 was used in the scenario that mimicked recoil when it was discharged.

Similarly, three other papers (Giacomantonio et al. 2020; VerPlanck 2020; Stenshol et al. 2024) used the MILO range simulator to train de-escalation tactics with or without the use of duty weapons. In these studies, the simulator responds to a participant’s reaction such as the drawing of their duty weapon and potentially using them. The simulation scenarios will play out based on the officers’ response. Zahabi’s et al. (2022) study involves the actual driving simulators to improve officers’ performance during operations. Using equipment simulators improves officers’ proficiency and on-the-job performance.

Lastly, five papers (13%) (Harris et al. 2021; Potts et al. 2022; Meenaghan et al. 2023; Kleygrewe et al. 2024a, 2024b) adopted head-mounted devices (HMDs) to train skills such as perceptual-cognitive, tactical procedures and movements, de-escalation techniques and communicative skills. Like other virtual simulations, VR allows adjustment and variation of virtual environments and use of multiple nonplaying characters. Sessions could also be tailored to specific trainee characteristics to be more effective (Kleygrewe et al. 2024b). These scenarios are usually not more than 15 minutes to manage physiological responses.

ANALYSIS

Trend in types of simulation exercises

An analysis was undertaken to compare the different types of simulation exercises done each year throughout the 10 years. Understandably, more live simulations were conducted within the first 6 years of study, from 2014 to 2019, where the technology for highly immersive virtual simulations had not yet matured. Correspondingly, as those were pre-COVID years, gathering people in large groups and having people act as role players were still considered the norm. Studies involving computer-based virtual simulations still occur steadily over the years, averaging around one paper each year in this 10-year period. This symbolises that computer-based simulations remains to be a stable and relatively easy-to-implement tool with shorter turnaround time whilst still meeting needs to train short, discrete tasks.

On the other hand, all of the papers that looked at highly immersive virtual simulation technologies took place from 2020, with the exception of Davies’s study in 2017. This coincides with the pandemic and demonstrates how the use of highly immersive simulation technology could potentially reduce the number of live exercises required for training. Figure 3 shows the trend of the different types of simulation exercises published in the 10-year period.

Figure 3

Types of simulation published each year

Despite technological advancements, live simulations involving role-players appear to be the preferred option for scenario-based simulation exercises. In such training exercises, officers often role-play as perpetrators or other stakeholders within the scenarios. Having officers who have encountered similar situations in their operational duties to role-play within scenarios adds realism to the exercise, thus increasing realism for officers who are being tested. In some other scenarios, situations are usually adapted from an actual incident and are non-fictional to retain the realism and fidelity of the situation, by stating upfront that the scenario is non-fictional, officers are more likely to accept its realism and immerse themselves in the exercise. Live simulations have been used to train officers to handle situations involving mental health crisis response (Alvarez 2020), hostage negotiation situations (van den Heuvel et al. 2014), domestic dispute cases (Cotard and Michinov 2018; Hine et al. 2018) and various UOF situations where it involves officers handcuffing the perpetrators (Armstrong et al. 2014; Hine et al. 2019). Where scenarios form part of officers’ assessment, role players often follow a specific script as they enact different behaviours. Depending on the officers’ responses, role players might escalate or de-escalate the situation. Such decisions remain within the script so that scenarios remain consistent to a certain degree and officers can be assessed appropriately.

In most cases, live simulation exercises are good for training officers to manage single training objectives that could be resolved by the officer participating in the exercise. In such simulations, the exercise ends with the officer resolving the incident without having to coordinate with other parties and agencies. Exercises and research involving such simulations often look at specific events, which could often be narrow, theoretically driven and removed from organisational contexts (Waring 2019). Often, they might also not be useful in uncovering complex relationships and coordination in real-world problems. In such instances, large-scale live simulations involving multiple teams might be more useful for studying multi-team systems. Live disaster simulations involving multiple government agencies would also be useful to assess the validity of national response plans in times of crisis.

Types of training within simulation training

For simulation training to be effective, the types of platforms and scenarios chosen must be fit for purpose. Training objectives of any simulation training should be considered and ascertained first before identifying the appropriate simulation platform to execute the training. An analysis of the different types of skills that were being trained in the simulation platform was tabulated and presented in Table 4. Only one study (Zahabi et al. 2022) utilised simulation to train multiple tasks. Particularly, the study used simulation to train driving skills of traffic police who are concurrently performing secondary tasks on a separate system. Integrating simulation systems with other training systems allows researchers to assess how different factors, such as increased cognitive load, heart rate, breathing rate, might impact officers’ job performance.

Table 4.

Types of skills trained within the simulation.

Types of skills	N	% of total articles (N = 37)
Daily operations	13	35
Decision making and cognitive skills	10	27
De-escalation training	7	19
Major incident management and inter-agency cooperation	5	14
Multiple tasks	2	5
Total	37	100%

Daily operations

Simulation training in the context of law enforcement is to prepare officers for day-to-day operations before being deployed. In probational training, simulation exercises allow officers to experience how situations might potentially unfold during deployment. This gives officers a chance to immerse in daily operations through the exercise. Common scenarios involving daily operations include incidents that might occur during police patrol, traffic control incidents and identity check of members of the public (Sjöberg 2014; Cotard and Michinov 2018). In some studies, simulation is also used for specific procedures such as drug recognition (Paquette and Prince 2021). As training to be operationally ready can be very general and broad, different types of simulation can be used for such training exercises. Six studies (Armstrong et al. 2014; Sjöberg 2014; Sjöberg et al. 2015; Johnsen et al. 2017; Cotard and Michinov 2018; Paquette and Prince 2021) deployed these simulation scenarios through live simulations, where other officers, trainers and even actors are deployed to act as role players in these scenarios. In some cases officers act as role players between different scenes and would have to prepare for their roles within the simulation for it to be effective (Sjöberg 2014; Sjöberg et al. 2015). Officers rotate between being assessed and being part of the role-playing team in such situations. As these officers are consciously aware of the presence of their trainers, whom they know are the ones evaluating their performance, officers have to have simulation competence, to know what needs to be done so that they can produce a meaningful scenario for their peers. When simulation competence is lacking, the simulation will be unclear and confusing, which would cause an adverse effect on the scenario, and not provide sufficient conditions for the experience to be deeper and have a higher quality learning experience (Sjöberg et al. 2015).

Training to be familiar with daily operations could also involve virtual tools such as video recordings and virtual desktop environments. Virtual desktop environments such as gaming simulators allow officers to explore operational areas and practice procedural knowledge (Söderström et al. 2019). In the case of the Söderström et al. (2019) study, the scenario within the virtual environment provided officers with the opportunity to discuss how to react to situations. Video recordings could also solicit similar stimulated recall opportunities to encourage officers to think through their responses after viewing clips of actual incidents through ground officers’ body-worn video camera (Phelps et al. 2016). Prompting officers to view their own recordings after a simulation also probes officers’ cognitive processes on why they had made certain decisions during the exercise.

More commonly in the last few years, the use of VR tools has been used to train officers in daily operations (Davies 2017; Potts et al. 2022; Kleygrewe et al. 2024b). Similar to scenarios that were conducted in live situations with role players, VR head-mounted headsets could also be used to train officers in managing day-to-day operations. In Kleygrewe et al. (2024b) study, the use of VR technology also allowed the tracking of physiological responses such as heart rate and physical activity. In recent years VR technology has been preferred to live simulations as it can allow the conduct of a practical test within a short period of time and can be done repeatedly with minimal manpower requirements (Potts et al. 2022).

De-escalation training

Tied closely to training to deal with daily operations, simulation is also often used for de-escalation training. De-escalation is a general term that encompasses all scenarios that require officers to resolve an ongoing situation within the exercise. Common de-escalation scenarios could involve officers de-escalating situations involving people with mental health issues, armed incidents and hostage situations. Seven such studies adopted the use of simulation for de-escalation training.

In daily operations, it is common for officers to meet with members of the public with mental health crises such as Alzheimer’s disease, schizophrenia, depression, etc. Officers who are able to effectively identify persons with mental health and have more knowledge about these individuals are able to better respond to these incidents (Helfgott et al. 2020). Two studies (Alvarez 2020; Helfgott et al. 2020) adopt the use of live role players to play out mental health scenarios and officers to refine their decision-making processes. Live role players provide the opportunity for the simulation to pause, go back or replay the scenario, so that officers could progressively build their tacit knowledge for future encounters. Similarly, live simulation was also used to train officers in hostage situations (van den Heuvel et al. 2014) by providing them with mental stimulation and the opportunity required to implement different strategies to de-escalate such situations. Lastly, the remaining de-escalation training study involves the use of the MILO Range simulator (VerPlanck 2020)

Virtual reality HMD has proven to be a valuable tool for training officers in de-escalation training for armed incidents. Using the highly immersive environment within a VR scenario, team training and tactical training have been made much easier as officers walked around in simulated virtual environments. In the Kleygrewe et al. (2023, 2024b) study, the VR system also allows other systems to be connected to it, such as tracking of an officers’ heart rate, and actions within the scenario, for further analysis to track learning efficacy to take place.

Decision-making and cognitive skills

Skills such as risk assessment, shooting accuracy, cognitive skills and UOF decision-making skills have been trained using simulations. Particularly, sound UOF decision-making skills are critical for police performance in critical incidents (Stenshol et al. 2024). Hence using simulation training is a helpful approach to train decision-making processes as there are audio-visual elements in a critical scenario simulation. Simulations contribute to professional development as carefully crafted exercises provide structured experiences in an environment that is designed to encourage flexible internalisation through the use of critical reflections (Phelps et al. 2016). Six studies used scenarios within the simulation to train UOF decision-making skills. These scenarios could start off with common scenarios faced during daily operations before branching out to training officers’ UOF decision-making skills (Cook et al. 2022). Such UOF decision-making skills are trained in simulations in the presence of civilians, where officers train their cognitive abilities in response inhibition, attentional deficits or impulsivity (Biggs et al. 2015). Three of these studies were conducted in virtual environments, such as the VirTra (Cook et al. 2022) and MILO range systems (Stenshol et al. 2024). The use of such highly immersive systems allows the simulation of recoil when the participants pull the trigger. Participants’ responses within simulated critical incidents gave participants a chance to make critical decisions, which allowed researchers to understand the difference between novice and expert decision-making (Stenshol et al. 2024). Di Nota et al. (2024) adopted both live and video simulations within the study to observe officers’ physiological responses and compare UOF performance between both modalities. Lastly, Hine et al. (2018, 2019) adopted live simulations to train the UOF decision-making skills. Choosing live simulations to train decision-making skills allows researchers to explore decision-making in naturalistic settings and for decision-making processes to be further examined (Hine et al. 2018). Officers go through these live simulations, followed by stimulated recall interviews, to understand their decision-making processes when de-escalating situations. In the study conducted by Liu et al. (2018), live simulations were also used to train officers’ shooting accuracy by simulating crowd-induced stress for officers in real-life hostage situations. The results showed that officers’ heart rates and anxiety scores were significantly reduced after the training with improved shooting performance.

Other than the UOF decision-making skills, simulations were also used to train perceptual, cognitive skills and risk assessment skills. Meenaghan et al. (2023) explored how risk assessment skills could be trained using XR technology through a HMD. The environment within the highly immersive conditions provides officers with the opportunity to ‘be in’ actual real-world homes rather than learning through role play. HMD devices were also explored to train perceptual-cognitive skills that are more prevalent in experienced officers (Harris et al. 2021; Scott et al. 2022). The immersive experience and the ability to move around the environment provide a similar experience as compared to an actual scene and have proven to be a viable method to train such skills.

Major incident management and inter-agency cooperation

Large-scale national disaster training requires the cooperation of multiple agencies and a huge amount of resources to coordinate personnel from different departments to come together for large-scale exercises. As law enforcement officers would most definitely be amongst the first to be deployed in civil emergencies, police officers would be likely to be involved in simulations that require joint agency cooperation. Such exercises are also hard to simulate in live environments as they disrupt normalcy and involve members of the public. There were a total of five studies (Wilson and Gosiewska 2014; Malizia 2016; Waring 2019; Korn et al. 2021; Mattingsdal et al. 2023) that used simulation exercises to train inter-agency cooperation. Of these, three studies (Wilson and Gosiewska 2014; Korn et al. 2021; Mattingsdal et al. 2023) were conducted through virtual platforms and tools, where injects are generated virtually and participants discuss their next course of action. Multi-agency exercises are often conducted to explore officers’ responses at the command level. Compared to ground operations training, decision-making at the command level requires commanders to synthesise information from different sources. Technological tools have allowed the creation of new injects on during the exercise, depending on the decisions of the participants. Whilst large scale interagency exercises are conducted to assess nation’s readiness towards disaster management, they also provide insights to how officers with different backgrounds engage in decision-making that transverse sectoral boundaries (Mattingsdal et al. 2023).

Technology has enabled interagency exercises at the command level to be conducted dynamically and suited to the ongoing decisions of participants within the exercise. Such exercises are not new and were previously conducted through table-top exercise (Malizia 2016) where injects are already pre-determined and any new injects or changes would be hard to create on the ‘fly’. Large-scale simulation exercises are a viable approach to test a nation’s response and resilience to major events and help to surface gaps in processes between inter-agencies, which would be difficult to surface if exercises are conducted within departments and in silos.

DISCUSSION

Understanding how law enforcement officers think and react during a crisis situation has been an area of interest for law enforcement agencies due to the high stakes natures of such situations. The decisions that officers make when responding to incidents could swing the outcomes of situations to extreme ends. Existing decision-making frameworks such as Naturalistic Decision Making (Lipshitz 1993) and Recognition-Primed Decision (RPD) model (Klein 1993) informs how experienced decision-makers are able to make accurate intuitive decisions that through pattern matching with past experience, an opportunity that novice decision-makers do not have until they have personally gone through similar experiences. However, officers do not have the luxury of going through crisis situations to train their decision-making skills. Decision-making is also a multi-faceted process where physical, physiological and psychological responses come into play when officers are put in a high-stake situation that requires prompt decisions. Simulation training has been the go-to approach to prepare law enforcement for crisis situations so that they can make sound judgements during incidents. This study has surfaced the preference for simulation training involving role-players and live actors up until recent years as immersive technology matures. As technology advances exponentially, simulations can now take place within the virtual space through a vast range of formats, varying from video-based systems to high-fidelity VR or AR systems (Di Nota et al. 2024). Virtual systems provide enhanced fidelity, flexibility and safety during training so that officers can be placed in highly stressful situations to improve their operational performance.

Whilst virtual simulation systems are possible alternatives to live role-player simulation training, interaction within virtual environments is still predominantly limited to audio-visual stimuli, which then hinders multi-sensory experiences (Giessing 2021). Whilst there are increasingly more studies to include other sensory feedback such as tactile stimuli through haptic feedback from the equipment, or olfactory stimuli such as heat from fire and smell, there remains limitations to fully recreate a sensory experience within the virtual environment. The efficacy of such training tools can also be difficult to ascertain as any form of feedback on the training is perceptive and acts as a proxy. In addition, highly immersive training systems are costly, and many agencies might not have the resources to own such sophisticated systems. Until officers’ skills are put to the test during actual incidents, the effectiveness of such simulation systems will remain tough to evaluate.

Technology has supported simulation training to be more immersive than before, with more tools and systems connected to each other in an attempt to make sense of the interrelationships between physical, physiological and decision-making characteristics during critical incidents. These insights could be useful in formulating effective learning experiences for pre-deployment officers so that they are well-prepared for any incidents that they might face during ground operations. However, this wealth of information to explain officers’ behaviour will only be useful towards officers’ learning if they are considered during the design of the learning activity. Placing officers through multiple highly immersive scenarios before deployment might create a preconceived notion of how ground operations might evolve. As officers build honed perceptual-cognitive skills through simulation training, excessive training might cause overconfidence and intuition based solely on virtual exercises that they have experienced. There will then be a risk of officers making premature decisions before fully assessing the situation, which has the potential to lead to more casualties during real-world operations. Simulation training has an effect on cognitive load. Repeated training for similar training outcomes might cause cognitive burnout from excessive redundant training (VerPlanck 2020). Hence, whilst law enforcement training is usually grounded on behavioural psychology, trainers designing simulation-based learning programmes also need to consider officers’ cognitive development, especially if they have only known traditional non-simulation-based learning experiences.

LIMITATIONS AND FUTURE RESEARCH

In this review, only empirical studies involving simulation-training exercises for law enforcement officers were reviewed and analysed. Papers that discussed design considerations of simulation training were not reviewed. These papers might contain useful information pertaining to the design and learning outcomes which the training aspires to achieve.

The search strings for this review were done in known databases for police training. However, simulation training covers a broad spectrum of disciplines across multiple databases. Hence, other papers pertaining to the subject might not have been uncovered, which would have limited the scope of the study.

Simulation training will continue to evolve with the development of new tools and technology. With generative artificial intelligence making its way into our lives and now mimicking human behaviour better than before, the field of simulation is continually escalating in complexity. Future research could explore how artificial intelligence can be part of scenario development, such that it evolves dynamically according to officers’ actions within the exercise. The interactions between participants and the system could provide further insights towards officers’ decision- and sense-making skills whilst reducing manpower costs during each training session.

CONCLUSION

The success of any training, simulation or otherwise, can only be determined when the practice during training leads to increased performance in corresponding real-world tasks (Harris et al. 2021). Fidelity within simulation allows training to be done in similar operational conditions with the hope that skills can be transferred easily during operations. Whilst virtual simulations promise to provide perceptual and physical fidelity in the future, the brain is finely tuned to detect asynchrony between senses (Spence and Squire 2003). Any disjoint in senses, such as visual cues not linking with officers’ movement in the virtual environment, will cause the scenario to be unnatural and might even cause negative physical effects. The brain also processes subtle facial muscle expressions, movements and speech patterns better during live exercises as compared to virtual interaction (Redcay et al. 2010). Hence, training with digital avatars might compromise officers’ visual perception which will have downstream effects on officers’ judgement, communication and decision-making skills (Di Nota et al. 2024).

This review highlights the broad spectrum of simulation tools and how they have been used for law enforcement training. It also emphasised that these tools should not be used in silo, they should be incorporated with other learning approaches, such as reflections and experiential learning. Officers should be given the opportunity to review what they have done within the simulation, reflect upon their decisions within the scenario and consider how they might improve performance. The use of simulation should be an international one, designed deliberately for different demographics and training outcomes. Only then will the training be fit for purpose and translate to improved work performances.

Conflict of interest. The authors conform no conflict of interest

Funding

The authors confirm no funding applied to this study

APPENDIX 1. SELECTED ARTICLES AND CODING TABLE

SN	Author and article information	Type of Simulation LS—Live Simulation C1—Computer based Interactive C2—Computer based Strategic C3—Computer based Video recording H1—Virtual Reality, Head-mounted devices H2—Highly Immersive and large screen simulators H3—Both H1 and H2 TT—Table-top exercises Hy—Hybrid exercise BG—Board Game	Training objectives T1—Daily operations T2—Decision making and cognitive skills T3—De-escalation training T4- Major incident management and inter-agency cooperation T5—Multiple Tasks
1	Alvarez (2020) Site: North America Profile: Police	LS	T3
2	Armstrong et al. (2014) Site: North America Profile: Police	H1	T1
3	Biggs et al. (2015) Site: Asia Site: North America Profile: Police	C1	T2
4	Cook et al. (2022) Site: North America Profile: Police	H2	T2
5	Cotard and Michinov (2018) Site: Europe Profile: Police	LS	T1
6	Davies and Krame (2024) Site: Australasia Organisation: Police	HY	T5
7	Davies (2017) Site: Australasia Organisation: Police	H2	T1
8	Di Nota et al. (2024) Site: North America Profile: Police	H3	T2
9	Giacomantonio et al. (2020) Site: North America Profile: Police	H2	T3
10	Harris et al. (2021) Site: Europe Profile: Police	H1	T2
11	Helfgott et al. (2020) Site: Europe Profile: Police	LS	T3
12	Hine et al. (2018) Site: Australasia Organisation: Police	LS	T2
13	Hine et al. (2019) Site: Australasia Organisation: Police	LS	T2
14	Johnsen et al. (2017) Site: Europe Profile: Police	LS	T1
15	Kleygrewe et al. (2023) Site: Europe Profile: Police	H1	T3
16	Kleygrewe et al. (2024) Site: Europe Profile: Police	H1	T3
17	Kleygrewe et al. (2024) Site: Europe Profile: Police	H1	T1
18	Korn et al. (2021) Site: North America Profile: Police	C2	T4
19	Liu et al (2018) Site: Asia Profile: Police	LS	T2
20	Malizia (2016) Site: Europe Profile: Multi agency	TT	T4
21	Mattingsdal et al. (2023) Site: Europe Profile: Multi-agency	C1	T4
22	Meenaghan et al. (2023) Site: Europe Profile: Police	H2	T2
23	Pacquette & Prince (2021) Site: Europe Profile: Police	LS	T1
24	Phelps et al. (2016) Site: Europe Profile: Police	C3	T1
25	Potts et al. (2022) Site: North America Profile: Police	H1	T1
26	Scott et al. (2022) Site: North America Profile: Police	C1	T2
27	Sjöberg (2014) Site: Europe Profile: Police	LS	T1
28	Sjöberg et al. (2015) Site: Europe Profile: Police	LS	D1
29	Söderström et al. (2019) Site: Europe Profile: Police	C1	T1
30	Söderström et al. (2022) Site: Europe Profile: Police	C3	T1
31	Stenshol et al. (2024) Site: Europe Profile: Police	H2	T2
32	Suss et al. (2014) Site: Unstated Profile: Police	C1	T1
33	van den Heuvel et al. (2014) Site: Europe Profile: Police	LS	T3
34	VerPlanck (2020) Site: North America Profile: Police	H2	T3
35	Waring (2019) Site: Europe Profile: Multi Agency	LS	T4
36	Wilson and Gosiewska (2014) Site: Europe Profile: Multi agency	C1	T4
37	Zahabi et al. (2022) Site: North America Profile: Police	H2	T5