Gamifying Traffic Incident Response: A Pilot Simulation Study at the Dubai Police Academy

Simulation in Law Enforcement Training

Over the past two decades, the use of serious games in professional education has expanded beyond military and aviation training to diverse civilian domains, including law enforcement (Akhgar et al., 2019) . Serious games integrate game mechanics with instructional objectives to create learner-centered environments that support skill acquisition through interactive decision making and real-time feedback. Compared to traditional instructional methods, such as lectures or paper-based exercises, simulations foster higher engagement and procedural fluency, which are essential in complex, high-stakes contexts, such as policing (De Gloria et al., 2014).

In policing, early simulation tools primarily address technical competencies, such as firearms judgment and vehicle operation. These applications are typically delivered using shoot/dont-shoot scenarios or basic driving modules. As simulation technology matured, its use expanded to encompass cognitive and procedural skills. One early example was the 3D traffic accident investigation simulation used by the Dubai Police, which enabled officers to analyze virtual crash scenes, collect evidence, and follow proper reporting procedures. Officers trained with the simulation outperformed those who received conventional instruction, highlighting the potential of serious games to enhance training effectiveness in law enforcement (BinSubaih et al., 2006).

Scenario-Based Learning and Procedural Fidelity

In recent years, research has focused on increasing the realism, variability, and transferability of simulation-based training. A systematic review by See and Davies (2024) documented a growing trend toward high-fidelity simulations that replicate real-world complexity in domains such as crisis de-escalation, tactical communication, and use of force decision-making. Virtual reality platforms, such as VirTra and MILO, are now used extensively by police agencies to simulate high-risk encounters. These platforms expose trainees to both routine and complex operational challenges in immersive environments, with dynamic consequences. Empirical studies have reported improvements in trainee confidence, preparedness, and situational awareness (Alshehhi et al., 2024). According to Wallentine (2021), officers who complete simulation training often demonstrate increased confidence in their field decision making. This outcome is closely linked to reduced use of force incidents and improved public safety performance.

Scenario-based learning continues to be a central strategy in law enforcement. Traditionally delivered through live role-plays and controlled field simulations, these methods allow trainees to practice managing incidents, such as traffic accidents, domestic disturbances, and felony stops. However, live exercises are costly, time-consuming, and difficult to standardize across different trainees and contexts. Digital simulations offer a scalable alternative that can reproduce a wide range of high-risk and high-stress scenarios with consistent instructional delivery (Caserman et al., 2018; Peeters et al., 2015). Many systems employ AI-driven virtual characters, branching narratives, and adaptive consequences to challenge users to make realistic decisions and respond to evolving situations.

These simulation designs are increasingly grounded in experiential learning theory (Kolb, 1984), which emphasizes iterative experience, reflection, conceptual understanding, and active experimentation in the learning process. Effective simulations are structured around a learning loop in which trainees make decisions, observe outcomes, reflect on performance, and improve subsequent attempts. Many platforms include replay functions and automated performance tracking aligned with Kolb’s learning cycle (Capuano & King, 2015; Peeters et al., 2015). In one example, failure to redirect traffic or summon medical support may trigger secondary accidents within the simulation, prompting learners to re-evaluate their procedural timing and priorities. This experiential model, widely applied in aviation and medicine, is becoming increasingly relevant to policing, where officers must manage complex, high-risk situations under pressure (Benda et al., 2020; Zhu et al., 2024).

Recent adaptations of Kolb’s model have extended its application to immersive technology-based environments, such as serious games and virtual simulations. Kolb and Kolb (2017) emphasized that digital experiential learning should preserve the structure of the cycle while also enhancing realism, learner autonomy, and contextual fidelity. In policing and related domains, simulations following these principles support active engagement, structured reflection, and immediate feedback. These conditions are critical for developing decision-making skills in complex, high-stake settings. Although direct applications of Kolb and Kolb (2017) in law enforcement are limited, studies on disaster training and healthcare education demonstrate that experiential designs can improve learner performance in environments that demand procedural accuracy and adaptive thinking (Bahaziq et al., 2023; Mitsuhara et al., 2019; Rholdon, Lemoine, Templet, & Stueben, 2020a, 2020b). These findings support the use of revised experiential learning models designed to improve operational readiness and legal compliance in police training simulations.

Gamification and Learner Motivation

Gamification, defined as the use of elements such as scoring, levels, and leaderboards, has emerged as a complementary strategy in simulation-based training. In policing contexts, gamification can improve learner engagement, focus, and retention, particularly among digital-native cadets (Alvear et al., 2020; Schöbel et al., 2022). Studies show that when implemented effectively, gamified simulations help trainees internalize both procedural knowledge and time-sensitive decision-making while maintaining cognitive load at levels similar to real-world stress conditions (Baah et al., 2024; Nair & Mathew, 2021). In the Dubai Police simulation, cadets received automated feedback and performance scores based on legal compliance, decision timing, and scene management accuracy. This feedback loop encourages mastery learning and supports self-assessments. However, the literature also warns that poorly designed gamification can reward speed over accuracy, or foster competition over cooperation. Instructional designers are advised to align game mechanics with educational goals, particularly those related to lawful conduct and decision quality (Gąsiorek, 2023; Schöbel et al., 2022).

Research Gaps and Study Purpose

Despite these advances, gaps remain in the empirical evaluation of simulation-based training, particularly in law enforcement contexts that require procedural fidelity. Much of the research has focused on tactical or communication-based training, with less emphasis on simulations that integrate legally accurate protocols and administrative procedures. This gap in empirical evaluation underscores the need for research that prioritizes legal and procedural fidelity in simulation-based police training, particularly for novice cadets. Integrating country-specific legal standards and replicating the high-stress environments of field operations is essential for ensuring training relevance and effectiveness. As Boembeke et al. (2022) emphasize, simulation scenarios often lack the depth needed to fully prepare trainees for the complex interplay of cognitive, legal, and procedural demands encountered in real-world policing. Addressing this oversight is crucial, as early-stage training forms the foundation of professional judgment and legal compliance in the field. Without such rigor, a simulation may fall short of its potential to reduce operational errors and improve public safety outcomes.

To address this gap, this study evaluates a gamified simulation model developed by the Virtual Technology Center and implemented at the Dubai Police Academy. The simulation focused on traffic incident response and incorporated branching narratives, procedural task execution, and real-time feedback based on UAE traffic law. Grounded in experiential learning and deliberate practice (Ericsson, 2008), this study investigated how simulation-based training affects decision-making accuracy, procedural compliance, and self-confidence among police cadets.

Simulation Model Design

The simulation was developed as a 3D interactive serious game, placing cadets in the role of first-responding traffic officers at the scene of a road accident. The development followed an iterative, user-centered design approach involving technical developers and subject matter experts from Dubai Police Academy. The simulation was built using a cross-platform development framework capable of rendering realistic urban environments and simulating complex interactions such as vehicle collisions, crowd behavior, and environmental hazards. The architecture supports deployment on PCs and multiple operating systems, including tablets and mobile platforms. Realistic environments and interaction dynamics are considered critical for user immersion and the transfer of learning in simulation-based training (Rehman & Huang, 2019; Zhou & Yang, 2022).

Grounded in experiential learning theory, the simulation aimed to provide cadets with authentic experience, structured reflection, and active experimentation in line with the experiential cycle (Kolb, 1984) and its application to immersive, technology-enhanced environments (Baah et al., 2024; Kolb & Kolb, 2017). The learning objectives were to ensure that cadets could (1) secure and manage a traffic accident scene, (2) apply UAE traffic laws and police procedures, (3) coordinate effectively with emergency services, and (4) accurately document the incident.

Scenarios follow a branching narrative structure, with multiple decision points and outcomes. Each session began with a briefing, such as “Two-car collision at Sheikh Zayed & 5th Street intersection”, followed by entry into a virtual scene. Decisions, such as whether to call an ambulance or initiate traffic control, influenced the scenario evolution. Errors or delays trigger in-simulation consequences resulting in worsening injuries and traffic build-up, reinforcing procedural accuracy, and real-time responsiveness (Caserman et al., 2023; Sharma, 2020).

Scenario content was developed in collaboration with Dubai Police subject matter experts based on the UAE Federal Traffic Law and internal standard operating procedures. The knowledge acquisition phase includes Standard Operating Procedures (SOP) reviews, expert interviews, and case studies of common cadet errors. Scenarios ranged from minor daytime collisions to multivehicle nighttime incidents. Logic programming allows the simulation to prompt or penalize cadets for protocol deviations such as delayed ambulance calls or improper evidence collection (Uddin et al., 2020; Verplanck, 2020).

Game Mechanics and Interface

The simulation was operated from a first-person perspective, with cadets using a keyboard and a mouse or built in game controller. Core features included a dialogue system for virtual interviews, an inventory of scene tools such as cones, radio communication with dispatch, and dynamic event timing to simulate real-world pressure.

As shown in Figure 1, the interface presents cadets with a realistic urban traffic scenario, in which they interact with scene elements, navigate procedural tasks, and manage unfolding events. The inclusion of Arabic User Interface and localized environmental features supports procedural fidelity and cultural immersion, consistent with best practices in user-centered virtual environments (Alsuwaidi & Almazrooei, 2025).OPEN IN VIEWER

The cadet actions were continuously logged and evaluated against expert-defined protocols. Correct decisions were rewarded with points; errors or omissions resulted in deduction. Post-session feedback included a performance summary with numerical scores and comments. Replay functions allow cadets to reflect on their actions and improve them iteratively (Toukiloglou & Xinogalos, 2023).

Scenarios incorporate variability such as weather, vehicle types, and time of day, to encourage adaptive learning and discourage memorization. Cadets are allowed to repeat scenarios to improve their performance (Bergeron, 2008). These features reflect the core principles of experiential learning in digital environments, where immediate feedback and repeated performance cycles help to translate procedural knowledge into field-ready competence (Kolb & Kolb, 2017).

Figure 2 presents a high-risk simulation scenario involving a fire, damaged vehicles, and public bystanders. This example demonstrates how environmental complexity and time pressure increase cognitive demands, requiring cadets to prioritize safety, communication, and evidence collection while maintaining procedural accuracy.OPEN IN VIEWER

Integration of Legal Protocols

Legal accuracy is the core design requirement. A senior legal advisor reviewed all scenario scripts to ensure alignment with UAE Federal Traffic Law. The simulation includes virtual procedures for accident severity classification and documentation. Cadets completed in-game reports using the official Dubai Police format, including drop-downs and free-text fields to log vehicle data, license numbers, and violation causes. Procedural missteps include prematurely moving vehicles, triggered score penalties, and real-time warnings. By embedding accurate legal workflows, the simulation reinforced procedural compliance and supported the transfer of learning to the field (Alshehhi et al., 2024).

Training Implementation

The simulation was piloted with 32 undergraduate students enrolled in the Bachelor of Science in Security and Criminology program, with a concentration in Traffic Studies, at the Dubai Police Academy. All participants had previously completed foundational classroom instruction on accident response. Participation in the study was voluntary, and informed consent was obtained from all cadets prior to their involvement. The data collected were anonymized, and participation had no impact on students’ academic performance or course grades. Ethical approval for the study was granted by the Dubai Police Academy Scientific and Ethics Committee.

Each cadet completed three simulation sessions:

1. Baseline Session: A scenario was completed without feedback to establish the pre-training performance.

Each scenario followed a distinct narrative structure and varied in complexity and environmental conditions. All cadets began with the same baseline scenario (Figure 1), followed by two differentiated training simulations and a final evaluation scenario (Figure 2). In total, cadets were exposed to four unique scenarios during the training sequence.

The training model was grounded in Kolb’s experiential learning cycle and Ericsson’s deliberate practice framework (Ericsson, 2008; Kolb, 1984). In line with recent adaptations of experiential learning to digital instructional contexts (Kolb & Kolb, 2017), the sessions emphasized learner-driven interaction, immediate feedback, and reflective iteration to promote procedural fluency (Beinicke & Muff, 2019; Stevens-Adams et al., 2010).

Performance metrics included:

• Decision accuracy : Percentage of optimal actions taken

Instructor observation forms complement system data and capture behaviors such as confidence, communication, and situational awareness (O’Neill et al., 2018).

Survey Instrument

Following the final scenario, cadets completed a post-training survey. The ten Likert-scale items were designed to assess key learning domains, including legal accuracy, confidence, realism, and decision prioritization. The items were adapted from validated simulation evaluation instruments (O’Neill et al., 2018; Tufts et al., 2021) and reviewed by Dubai Police experts to ensure contextual relevance. The survey was delivered in Modern Standard Arabic. Items were translated from English using a forward-backward translation process and reviewed by bilingual instructors to ensure clarity and conceptual equivalence.

A 5-point scale was used (1 = Strongly Disagree, 5 = Strongly Agree). Two open-ended items captured the participants’ perceptions of the challenges and the training’s overall instructional value. Appendix A provides the complete survey and its alignment with the learning objectives.

Responses to the open-ended items were analyzed using a basic thematic analysis. All comments were reviewed and manually coded by the lead researcher who grouped recurring ideas and phrases into emergent categories. The process was inductive and aimed to identify shared perceptions across participants while acknowledging the limitations of the small qualitative dataset.

Data Analysis

To evaluate the training impact, both performance metrics and participant feedback were analyzed using a mixed-methods approach. Quantitative performance scores from the simulation were compared using paired-sample t-tests (α = 0.05) and effect sizes were calculated using Cohen’s d to determine the magnitude of change. To analyze self-assessment responses, Wilcoxon signed-rank tests were applied given the ordinal nature of the Likert scale. Descriptive statistics summarized all survey responses, and open-ended feedback was thematically coded to identify patterns in cadet perceptions and reported challenges (Vatral et al., 2022; Zayapragassarazan et al., 2023).

This combination of system logs, instructor observations, and subjective feedback provided a comprehensive picture of how simulation influenced decision-making, procedural compliance, and learner confidence (Attoe et al., 2017; Davies & Heysmand, 2019).

Performance Improvement

The evaluation showed substantial improvements in the cadets’ ability to manage traffic incident scenarios following simulation-based training. Table 1 summarizes the mean performance scores across 32 cadets before and after the intervention.

• Decision-making accuracy improved from 72.3% to 89.7% (+17.4 percentage points), reflecting better responses to branching decision points under pressure.

OPEN IN VIEWER

Table 1.

Cadet Performance Metrics before and after Simulation Training.

Performance Metric	Pre-Training Mean (SD)	Post-Training Mean (SD)	Improvement
Decision-making accuracy	72.3% (±12.4)	89.7% (±5.8)	+17.4%
Procedural compliance	65.1% (±15.3)	94.0% (±4.1)	+28.9%
Report documentation quality	68.4% (±14.8)	85.2% (±9.6)	+16.8%
Overall performance score	70.0%	90.5%	+20.5%

Note. “Pre-Training” refers to baseline scenario performance; “Post-Training” reflects final evaluation after simulation use. All improvements are statistically significant at p < .01.

All the improvements were statistically significant. Paired-sample t-tests showed a strong significance (p < .001 for accuracy and compliance; p = .002 for reporting). The effect sizes were large (Cohen’s d > 1.0), indicating statistical and practical significance. Notably, all cadets improved; the largest gains were among those with lower baseline scores, consistent with the findings that novices benefit substantially from structured simulation training (Beinicke & Muff, 2019).

The most pronounced gain occurred in procedural compliance, suggesting that cadets internalized the correct sequence of actions such as securing the scene, collecting evidence, and completing reports. During the debriefs, cadets reported developing mental checklists reinforced by simulation feedback loops. Instructors’ observations corroborated these improvements. Cadets who initially focused on a single task, later demonstrated the ability to manage multiple tasks simultaneously, such as coordinating medical assistance while managing traffic flows. These behavioral changes were also reflected in the simulation log data and confirmed through structured instructor assessments (See & Davies, 2024).

Survey Results and Trainee Perceptions

Post-training survey responses offered additional insight into cadet confidence, perceived realism, and instructional value. Pre-training scores were collected retrospectively following the final session, to reduce response burden and capture perceived growth. These ratings reflect cadets’ recalled perceptions rather than real-time self-assessments. Table 2 presents a summary of the key Likert scale responses based on the 10-item instrument described in Appendix A.

OPEN IN VIEWER

Table 2.

Trainee Self-Ratings and Perceptions (Likert Scale 1–5).

Survey Statement	Pre-Training a	Post-Training
Confidence in managing a traffic accident scenario	3.2 (±0.7)	4.5 (±0.5) b
Understanding of traffic incident procedures	3.4 (±0.6)	4.7 (±0.4) b
The simulation scenarios felt realistic and immersive	–	4.6 (±0.5)
The simulation helped improve my decision-making under pressure	–	4.7 (±0.4)
I Would recommend incorporating this simulation into regular training	–	4.8 (±0.3)

As the pre-training ratings were collected retrospectively, they may be subject to the 'hello -goodbye’ effect (Salkind, 2006), where participants’ recalled perceptions of their baseline skills are influenced by their post-training experience, potentially inflating perceived improvement.

Responses to the two open-ended items were analyzed using a basic thematic analysis. Three recurring themes emerged across cadet feedback.

1. Realism and immersion

Cadets described the simulation as lifelike, relevant, and reflective of real field challenges. Many highlighted that the experience helped them internalize a “mental playbook” to respond to future incidents. One participant wrote,

“Initially I was unsure what to do first. Now, I feel I have a clear game plan if this happens in real life.

Others valued the ability to make mistakes in a safe environment, noting that,

“The game let me make mistakes and learn from them.

Trainees also offered constructive suggestions, including requests for expanded scenario types (e.g., pedestrian accidents), optional tutorials, and more flexible AI dialogue. These recommendations were recorded for future iterations (Choudhary & Jalan, 2022).

Instructor feedback was aligned with trainees’ perceptions. Trainers observed unusually high levels of focus and engagement even during peer observation. Some cadets voluntarily repeated the scenarios to improve their scores, suggesting sustained motivation. Instructors also reported that cadet decisions closely aligned with real-world procedural expectations. For example, when one cadet prematurely moved a vehicle, the simulation flagged the action, prompting group discussion, and reinforcing the proper field protocol.

Bridging the Gap Between Theory and Practice

This pilot study provides preliminary evidence that gamified simulation training may enhance procedural readiness among police cadets in localized traffic scenarios. The observed improvement in decision-making accuracy, procedural compliance, and confidence underscores the effectiveness of simulations in addressing the persistent gap between classroom instruction and operational performance (Alshehhi et al., 2024; Xue et al., 2021). Traditional police training often emphasizes memorization and written assessments, which may leave cadets ill-equipped to manage the pace, pressure, and complexity of actual field scenarios (Kienitz et al., 2024).

Our findings align with recent research on serious games as experiential tools that shift learning from “what to know” toward “how to act.” This approach fosters immersion, reflection, and practical competence (Facchino et al., 2025). Many cadets in our study initially expressed uncertainty about protocol application, despite prior instruction. However, after repeated simulation exposure, they demonstrated improved procedural fluency under time pressure, supporting the argument that action-oriented feedback-rich training enhances operational readiness.

Procedural Compliance and Experiential Learning

The nearly 30-percentage-point gain in procedural compliance highlights the capacity of the simulation to instill the sequential execution of police protocols. Real-world traffic incidents do not permit instructional pauses or repeated trials, yet the simulation environment allows for both, offering implicit feedback through simulated consequences, such as victim deterioration or scene escalation. This feedback loop promotes learning through reflection and adaptation, which are core components of Kolb’s experiential learning cycle (Kolb, 1984). Recent extensions of the model emphasize the importance of immersive, realistic training environments that support learner agency and contextual fidelity, which are features that are central to the simulation’s instructional design (Kolb & Kolb, 2017). These results also reinforce recent findings that repetition and real-time error correction improve protocol retention in public safety education (Calandra et al., 2022; Rholdon et al., 2020a, 2020b).

The cadets remained highly engaged throughout the training, further validating the model’s instructional design. Simulation-based learning has been shown to sustain attention and motivation by placing learners in dynamic decision-driven scenarios that mirror field complexity (Alshehhi et al., 2024; Toukiloglou & Xinogalos, 2023). In this study, repeated exposure to realistic, high-pressure situations encouraged cadets to internalize procedural sequences while developing decision-making flexibility. Their observed gains reflect not only technical skill acquisition but also enhanced cognitive readiness, defined as the ability to apply structured judgment under evolving conditions,a critical competency in law enforcement training (See & Davies, 2024).

Confidence and Field Readiness

Increased cadet confidence, as measured by both quantitative self-ratings and open-ended feedback, was especially significant. Confidence is not merely an affective outcome; it directly shapes performance under stressful conditions. Simulation-based training has been shown to improve confidence by reinforcing procedural memory and reducing hesitation in critical situations (Lambert et al., 2023; Tufts et al., 2021). Cadets in this study reported a transition from initial uncertainty to having a structured “mental script” for field responses. This internalization may improve decision speed, reduce procedural errors, and enhance public safety.

Engagement and Learning Culture

The gamified elements of the simulation contributed to a self-sustaining learning environment. Cadets voluntarily repeat scenarios to improve their scores and share strategies with their peers. This form of peer-driven engagement is uncommon in traditional training and demonstrates the potential of gamification to increase time-on-task and deepen learning (Rajagopal et al., 2020; Toukiloglou & Xinogalos, 2023). Crucially, scoring and competitive elements were aligned with procedural accuracy rather than speed, ensuring that engagement enhanced instructional goals, rather than diverting attention from them.

Protocol Fidelity and Instructional Alignment

Importantly, the realism of the simulation did not compromise the procedural integrity. The co-design process involves close collaboration between simulation developers, legal experts, and police trainers. As a result, the scenarios accurately reflected the UAE traffic law and Dubai Police protocols. Instructors confirmed that cadet performance in simulations aligned closely with field expectations, and feedback on procedural errors (e.g., moving a vehicle prematurely) mirrored real-world training discussions. This instructional fidelity not only enhanced the transfer of learning but also fostered instructor trust, which is a key factor in the institutional acceptance of new training tools (Boembeke et al., 2022; See & Davies, 2024).