Does Peer Influence Impact Workplace Performance and Safety Practice in a Virtual Environment?
Project Overview
Background
Reports have shown that peers influence decision-making even in the workplace. While unmanaged peer influence can cause spread of prosocial behaviors, it can also spread unwanted behaviors.
In an effort to improve team dynamics and performance, this study aimed at assessing peer influence on a person's productivity and safety practice.
Objectives
-
Assess if peer influence impacts productivity and safety practice.
Outcome
-
Validated a theoretical model of peer effects in the workplace.
-
Demonstrated measurable behavioral changes through peer information design
-
Identified strategies for balancing. productivity and safety in workplace feedback systems.
-
Produced a usable, validated virtual simulation tool for future research and training.
Team
-
My role: Doctoral researcher, Project manager, Lead software developer, UX researcher.
-
Team members:
-
2 project supervisors
-
1 UX researcher
-
Tools Used in this Project
Unity3d
For simulation game development
SQL
Designing the database that stored user behavior
Python
For data analysis (data cleaning)
Minitab
For data analysis (EDA and hypothesis testing)
How Did I Solve the Problem?
Methodology
Literature Review
Reviewed 103 peer-reviewed articles to identify why peer influence occurs (motivation), how it occurs (mechanism), and its impacts in workplace settings.
Software Development
Built a Unity 3D commercial kitchen simulation that displayed virtual coworkers’ productivity and safety scores to operationalize peer effects via social comparison and learning.
1:1 Interviews (usability testing)
Conducted individual interviews to validate the simulation’s ecological realism, clarity of instructions, and effectiveness of peer-information design.
Comparative Analysis
Had 112 participants complete baseline and treatment (peer-info) conditions and measured changes in productivity and safety behaviors under different peer-score displays.
Literature Review
Peer Influence Conceptual Model
What did I do?
I conducted a systematic review of 103 peer-reviewed articles on peer influence in workplace settings. The goal was to synthesize insights across three dimensions:
-
Why peer influence occurs (motivation)
-
How it occurs (mechanism)
-
What impact it has on workplace outcomes
I focused on extracting theoretical frameworks, empirical findings, and proposed models relevant to these themes.
What were the results?
The review identified two dominant mechanisms of peer influence:
-
Social comparison – Individuals adjust their behavior based on how they compare themselves to others.
-
Social learning – Individuals imitate behaviors observed in peers, especially when those behaviors yield positive outcomes.
Peer influence requires that a peer’s behavior is observable and perceived by others. Once perceived, the receiver evaluates their own behavior in light of that information. If motivated, this can lead to behavioral change.
Findings on the impact of peer influence were mixed—some studies reported performance improvements, while others noted negative effects such as conformity pressure or distraction. Notably, none of the reviewed studies directly examined peer influence on safety behavior.
What challenges did I face?
-
Inconsistent definitions of "peer influence" made comparison across studies difficult.
-
Many studies focused on outcomes without clearly identifying the mechanisms involved.
-
A lack of direct research linking peer influence to safety behavior required extrapolation from related domains.
Despite these challenges, I synthesized the literature into a theoretical model outlining how peer influence operates in workplace settings. This model became the foundation for a virtual simulation I later developed to study peer effects on productivity and safety.
Software Development
What did I do?
I developed a browser-based simulation using Unity 3D to study peer effects on workplace performance in a realistic commercial kitchen environment. Users completed food prep tasks while being observed by virtual coworkers whose performance (productivity and safety) was visible in real time.
The simulation was designed to trigger peer influence through social comparison and competition, based on the theoretical model I built during my literature review. I implemented real-time feedback on user and peer performance to drive behavioral adaptation.
I also built a MySQL database with PHP integration to securely store user data, track session performance, and manage experimental conditions.
What were the results?
The result was a fully functional simulation platform with:
-
A dynamic UI showing real-time productivity and safety scores for users and virtual peers
-
Customizable peer profiles (e.g., high-safety, low-productivity) based on experimental needs
-
A task management system assigning equal-difficulty prep tasks, tracking completion time, and safety compliance
-
Behavior logging for later analysis of performance and decision-making
-
Integrated visual cues and leaderboard displays to simulate peer pressure
What challenges did I face?
-
Designing believable virtual coworkers without overwhelming the user required careful tuning of animations and feedback frequency
-
Finding the right balance of peer feedback was critical—too much broke immersion; too little reduced peer effect triggers
-
Ensuring cross-browser stability and smooth performance involved extensive UI testing and resource optimization
1:1 Interviews (usability testing)
What did I do?
I conducted Alpha and Beta testing through 1:1 interviews with 13 participants to evaluate the simulation’s ecological validity, instructional clarity, and peer effect mechanisms. Participants engaged with the simulation while I observed gameplay and asked open-ended questions.
Key objectives included assessing:
-
Ecological validity – How realistic the simulation felt
-
Ease of understanding – Clarity of instructions and task flow
-
Design effectiveness – Whether peer effects were noticed and understood
What were the results?
-
Alpha test findings:
-
8/13 participants said instructions were too long and easily forgotten
-
10/13 participants ignored the leaderboard due to task focus
-
-
Beta test improvements:
-
Instructions were shortened and verbally emphasized before gameplay
-
Introduced 3-second breaks between tasks and repositioned leaderboards to the bottom of the screen for visibility
-
-
Beta test outcomes:
-
Participants found the kitchen environment realistic, confirming ecological validity
-
Instructions were clear and easy to follow, with minor wording suggestions
-
11/13 participants reported feeling motivated to outperform virtual coworkers, confirming that peer influence mechanisms were effective
-
These results validated that the simulation successfully triggered social comparison and learning, supporting its use as a tool for studying peer effects on safety and productivity.
What challenges did I face?
-
Balancing experimental control with user feedback—many participants suggested enhancements (e.g., voice prompts) that risked adding confounds
-
Ensuring feedback remained focused on research objectives, not just game aesthetics or entertainment value
Despite these constraints, the interviews were critical for refining the design and confirming that the simulation faithfully represented real-world peer dynamics.
What did I do?
I recruited 112 participants using random sampling and snowballing methods, including flyers, class announcements, and student research platforms, to test how peer performance visibility influences workplace behavior in a virtual simulation.
A participants performance was measured in terms of their:
-
Safety practice- Sratio (Number of times safety tool was correctly used/Number of tasks completed)
-
Productivity - Tavg (Time to complete a task/Number of tasks completed)
Each participant first completed a baseline session with no peer information. They were then exposed to treatment conditions displaying virtual coworkers' productivity scores, safety scores, or both.
What were the results?
Using ANOVA and Post-hoc Tukey's test the results suggested:
-
Combined low levels of peers productivity + safety cues led to the greatest gains in both safety and productivity, proving these goals are not mutually exclusive.
-
Exposure to high productivity alone led to reduced safety compliance—suggesting corner-cutting, however pairing that information with safety information reduced the negative impact on safety compliance.
-
The study showed that unbalanced performance feedback (e.g., focusing only on speed) can unintentionally promote risky behavior, however when paired with safety behavior information reduces this negative impact.
-
This research showed that exposure to peer performance can induce pressures which can lead to safety behavior avoidance, however when peer safety information is also shown, the egative impact on safety compliance is reduced.
These findings demonstrate that carefully designed peer feedback systems can encourage both efficiency and safe work practices—with real implications for workplace design, performance monitoring, and organizational safety culture.
What challenges did I face?
-
Balancing realism and experimental control, especially when simulating peer behavior.
-
Managing the natural tradeoff between safety and productivity, and preventing cognitive overload from competing cues.
Comparative Analysis
