Equitable Design in Motion: My RIDE Fellowship at UMass Amherst

Equitable Design in Motion: My RIDE Fellowship at UMass Amherst
Virtual Reality for Safer Bicycle Infrastructure
Thema Green — Human Factors & UX Researcher
This presentation details my fellowship experience in the NSF-funded Research for Inclusivity and Driving Equity (RIDE) REU program, where I contributed to developing virtual reality environments to study bicycle infrastructure design and safety. My work bridged technical development with human-centered design to create more equitable transportation systems.
The RIDE REU Program Context
The Research for Inclusivity and Driving Equity (RIDE) program at UMass Amherst is a 9-week NSF-funded Research Experience for Undergraduates focused on inclusive and accessible design approaches. Located within the Human Performance Lab (HPL), the program bridges computer science, human factors, and transportation engineering.
My project focused on developing and utilizing a VR bicycle simulator to test infrastructure safety designs, particularly comparing different intersection treatments. The program emphasizes:
Interdisciplinary collaboration between engineering, psychology, and design
Hands-on research experience with advanced technologies
Mentorship from faculty and graduate students
Professional development workshops and seminars
The HPL's mission centers on creating safer transportation infrastructure through empirical research. This lab specifically focuses on developing virtual reality environments that can test real-world designs before implementation, allowing researchers to collect behavioral data that would be difficult or dangerous to gather in actual traffic situations.
The RIDE program serves as a pipeline for underrepresented students into graduate studies in human factors, transportation engineering, and virtual reality research.
Why Bicyclist Safety Matters
The urgency behind this research becomes clear when examining the data from Massachusetts:
904
Bicycle Crashes
Documented bicycle-vehicle incidents in Massachusetts during 2022
8
Cyclist Fatalities
Deaths resulting from bicycle-vehicle collisions in the state
2.6%
Of Total Roadway Deaths
Proportion of total traffic fatalities represented by cyclists
Danger Zones
Intersections represent the most hazardous locations for cyclists, accounting for approximately 75% of all bicycle-vehicle collisions. The design of these intersections—particularly whether they include dedicated cycling infrastructure—significantly impacts safety outcomes.
Research has consistently shown that perceptions of safety vary widely based on cyclist experience level, age, gender, and other demographic factors. Less experienced riders, women, older adults, and members of marginalized communities often report higher levels of stress and anxiety when navigating urban cycling infrastructure.
This equity gap in perceived safety represents a significant barrier to increasing cycling adoption among diverse populations, making it a critical area for human factors research.
Bicycle crash hotspots in Massachusetts, with notable concentrations in urban centers
My Role in the Human Performance Lab
Literature Review
Analyzed 30+ studies on bicycle infrastructure
Focused on bike boxes vs. sharrows effectiveness
Identified gaps in research for novice riders
Applied equity lens to existing literature
Unity3D Development
Created immersive VR intersection environments
Programmed realistic traffic behaviors
Conducted and led usability studies to gather real user feedback
Optimized for VR performance (90fps+)
Integrated eye-tracking capabilities
Data Collection
Recruited and screened participants
Administered pre/post surveys
Operated VR equipment and troubleshooting
Calibrated eye-tracking systems
Graduate-Level Integration
Weekly meetings with lab PI and grad students
Iterative research design improvements
Accountable for project milestones
Contributed to lab publications
My position as a Research Assistant in the Human Performance Lab involved considerable responsibility across the entire research process. While the official project description emphasized Unity3D development and literature review contributions, my actual role expanded significantly to include data collection, experimental design, and analysis work.
This comprehensive involvement allowed me to experience the full lifecycle of human factors research, from initial concept development through to data interpretation and reporting. 
I functioned effectively as a graduate-level researcher, despite my undergraduate status, taking ownership of key project components and contributing meaningfully to the lab's research objectives.
The Unity3D development environment where I created the VR bicycle simulation, showing both bike box and sharrow intersection designs that participants would experience.
The Research Process
Literature Review
Identified research gaps in bicycle infrastructure studies, particularly regarding novice riders and diverse populations. Focused on comparing bike boxes (dedicated spaces at intersections) versus sharrows (shared lane markings).
Simulation Development
Created realistic VR intersections in Unity3D featuring both infrastructure types. Optimized 3D assets for VR performance, implemented traffic logic with realistic vehicle behaviors, and integrated physics for natural cycling movement.
Experimental Design
Developed a within-subjects design where participants experienced both infrastructure types. Created protocols for consistent data collection, including pre/post surveys, behavioral measures, and eye-tracking procedures.
Data Collection
Facilitated participant sessions, managing VR equipment and eye-tracking calibration. Collected data from diverse participants across experience levels, ensuring proper execution of the experimental protocol.
Analysis
Used R/RStudio to perform statistical analysis (ANOVA, t-tests) on quantitative data. Created visualizations of eye-tracking data including heatmaps and scan paths to reveal attentional strategies.
Reporting
Developed research posters and presentations for the REU symposium. Contributed to drafting journal article sections describing methodology and preliminary findings.
This end-to-end research process mirrored graduate-level work, giving me comprehensive experience in human factors research methodology. Throughout each phase, I maintained regular meetings with my faculty mentor and graduate student advisors, ensuring that my work aligned with established research practices while incorporating my own insights and innovations.
Challenges & Iteration
Technical Challenges
Unity Performance Issues
Initial simulation builds crashed frequently due to polygon-heavy assets and complex physics calculations. I systematically reduced polygon counts, optimized lighting, and implemented level-of-detail (LOD) systems to maintain visual fidelity while improving performance.
Eye-Tracking Calibration
Eye-tracking in VR presented unique challenges with calibration accuracy. I redesigned the calibration process to include verification steps and developed a recalibration protocol that reduced data loss by approximately 40% in subsequent testing sessions.
Research Design Challenges
IRB Revisions
Initial IRB submission required multiple revisions to address VR safety concerns, particularly regarding simulator sickness. I developed enhanced screening protocols and introduced gradual VR exposure techniques that successfully reduced discomfort reports by 65%.
Survey Refinement
Pilot testing revealed confusion with technical terminology in surveys. I rewrote questions using Human Factors and Ergonomics Society guidelines for clarity, conducted cognitive interviews to validate understanding, and implemented validated scales where appropriate.
These challenges exemplify the iterative nature of human factors research. Each obstacle presented an opportunity to develop problem-solving skills and improve the overall research design. The experience taught me that flexibility and persistence are essential qualities for successful research implementation.
"The most valuable research skills I developed weren't technical—they were adaptability and systematic problem-solving. Learning to view setbacks as opportunities for improvement transformed my approach to research."
Data & Early Findings
Preliminary Results
Our pilot study with 24 participants (12 experienced cyclists, 12 novices) revealed several significant patterns:
Safety Perception: Bike boxes were perceived as significantly safer than sharrows across all participants (p < .05), with a more pronounced effect among novice riders
Performance Differences: Experienced cyclists navigated both infrastructure types more quickly and reported higher confidence levels
Anxiety Measures: Novice cyclists showed elevated heart rate and self-reported anxiety in sharrow conditions compared to bike box scenarios
Gender Differences: Female participants reported higher stress levels in both conditions, but the gap narrowed significantly in the bike box condition
These findings align with previous research while providing new insights into the specific mechanisms through which infrastructure design affects cyclist comfort and behavior.
Average Cyclist's Safety Ratings in Bike Box vs. Sharrow Environments
Mean safety ratings (1-10 scale) by cyclist experience and infrastructure type
Eye-tracking heatmap showing attentional focus differences between novice (left) and experienced (right) cyclists
Eye-Tracking Insights
Perhaps the most revealing findings came from our eye-tracking data, which demonstrated distinct attentional strategies:
Novice Cyclists
Exhibited broader scanning patterns, frequently shifting attention between multiple elements (signals, road markings, approaching vehicles). Spent 43% more time looking at peripheral areas and potential threats.
Experienced Cyclists
Demonstrated more focused gaze patterns, quickly identifying and fixating on relevant infrastructure cues and potential conflict points. Spent 67% more time focusing on lane positioning and direct path of travel.
Professional Development & Community Engagement
Weekly Research Seminars
Research ethics in human subjects testing
Reproducibility in experimental research
Statistical methods for behavioral data
Scientific communication and visualization
Professional Workshops
Scientific abstract writing
Research poster design and presentation
Graduate school application strategies
Research funding and grant writing basics
Community Engagement
Volunteered at Amherst Survival Center
Participated in local cycling advocacy meetings
Field trip to Volpe Transportation Center
Connected research to community needs
Graduate-Level Research Culture
Beyond the structured activities, the most valuable aspect of the fellowship was immersion in a graduate-level research environment. The lab culture emphasized:
Accountability through regular progress reports and milestone deadlines
Intellectual curiosity that extended beyond assigned tasks to broader research questions
Collaborative problem-solving with graduate students and faculty
Constructive critique through feedback sessions and peer review
This environment allowed me to develop not just technical skills, but also the professional mindset and habits essential for success in graduate-level research. I learned to think critically about research questions, anticipate methodological challenges, and communicate findings effectively to diverse audiences.
"The fellowship transformed my understanding of what research involves—it's not just about finding answers, but asking better questions. The graduate students modeled a level of curiosity and rigor that I've tried to incorporate into my own approach."
The combination of structured professional development activities and informal mentorship created a comprehensive learning experience that prepared me for future graduate studies and research careers in human factors, transportation engineering, or virtual reality development.
Skills Development & Growth
Technical Skills
Unity3D VR Programming
C# scripting for interactive environments
VR optimization techniques
Physics simulation for realistic cycling
Asset management and scene construction
Eye-Tracking Integration
Tobii eye-tracker implementation
Calibration procedures
Data collection and parsing
Fixation and saccade analysis
Data Analysis
R/RStudio statistical analysis
ANOVA and t-test implementation
Data visualization techniques
Experimental design principles
Human Factors Skills
Experimental Design
Within-subjects testing protocols
Controlling for confounding variables
Counterbalancing techniques
Pilot testing methodologies
Ethics & IRB Procedures
Human subjects protection
Informed consent processes
Risk assessment and mitigation
Data privacy and security
Usability Testing
VR-specific usability considerations
Survey design and validation
Think-aloud protocols
Inclusive design principles
Professional Skills
Scientific Communication
Research poster creation
Abstract and paper writing
Presentation of technical content
Visual communication of data
Research Collaboration
Interdisciplinary team communication
Project management techniques
Constructive feedback exchange
Version control and documentation
Graduate-Level Work Habits
Independent problem-solving
Research literature synthesis
Iterative improvement processes
Time management for complex projects
This comprehensive skill development aligned perfectly with the lab's expectations while exceeding my own learning goals. The technical, methodological, and professional capabilities I developed provide a strong foundation for future graduate studies and research careers in human factors, transportation engineering, or virtual reality development.
From Simulation to Safer Streets: Reflection & Future Directions
"This fellowship was the turning point where I became a human factors researcher. I learned to code VR environments, analyze eye-tracking data, and think critically about equitable design; but more importantly, I learned the rhythm of real research: iteration, mentorship, and impact."
The RIDE REU program at UMass Amherst's Human Performance Lab provided me with transformative experiences that bridged technical development with human-centered design approaches. By contributing to research on bicycle infrastructure safety, I developed practical skills while advancing knowledge that can lead to more equitable transportation systems.
The most valuable insight I gained was understanding how experience level fundamentally changes how people perceive and interact with infrastructure. This insight has profound implications for designing inclusive transportation systems that work for everyone, not just confident cyclists.
Next Steps
This experience has solidified my determination in pursuing a career in human factors engineering with a focus on human-computer interactions. I aim to continue exploring how innovative research methodologies can bridge the gap between technical capabilities and human needs, ultimately creating safer and more accessible environments for diverse users.
The bike box design that showed improved safety perceptions across all participant groups in our study
Thank You
Dr. Anuj Pradhan, Faculty Mentor
Manoj Paari & Apoorva Hungard, Graduate Student Mentors
The Human Performance Lab team
NSF RIDE REU Program
UMass Amherst College of Engineering
Study participants who shared their experiences
For more information about this research or the RIDE REU Program, please contact me at [email protected]
Made with