Design System for VR Training simulations

// virtual reality training simulation

We were tasked with developing a design system for VR training that makes simulations more impactful for learners and more efficient for production teams. In addition to modular templates, we developed reusable interaction patterns that served clear learning outcomes. The system directly responds to pain points from legacy sims, where learners often felt stuck, frustrated, or unclear about their performance. By emphasizing agency, feedback, and meaningful choice, it transforms training into a more engaging and effective training framework, helping learners understand not just what to do, but why it matters.

Play movie icon
Case study summary video (2:45)

Company. TransfrVR

Timeframe. 2 months

Responsibilities.

  • Served as Creative Director and prototyper for the design system initiative
  • Partnered with the Learning Intelligence team to translate user testing data into actionable design insights
  • Collaborated with the Lead Instructional Designer to prototype and iterate training sims
  • Facilitated cross-functional design workshops and alignment sessions
  • Secured stakeholder feedback and buy-in to guide next-phase development

Limitations of Legacy Training Sims

We audited our legacy training sims, playing them firsthand and gathering feedback from product, learning, research, and customers to pinpoint what worked and where they fell short.

Linear and restrictive user flows

  • Early sims forced a single “correct” sequence, limiting realism and engagement.
  • The user had no opportunity for meaningful choice and learners couldn’t explore alternate paths or make authentic mistakes.
  • Tasks that could be done in any order in real life were locked into one path. Users who deviated couldn’t progress and were penalized unfairly.
  • These sims were created over four years ago during VR’s experimental phase, before training and UX best practices were established. This led to “showstoppers” where users knew the correct action but couldn’t perform it in VR, preventing them from progressing in the simulation and causing drop-offs that reduced renewal rates.

Weak Learning Feedback and Assessment

  • Errors were tied to idle time rather than real decisions, making it unclear whether failures reflected content knowledge or VR usability.
  • The user had no opportunity for meaningful choice and learners couldn’t explore alternate paths or make authentic mistakes.
  • Lack of immediate correction meant users rarely understood why they were wrong, reducing retention and transfer to real-world tasks.
  • At the end of the simulations, users were given scores, but were never told what mistakes were made, removing the chance for the learner to understand or remediate errors.

Research Insights that Shaped the Design System

In collaboration with the Learning Intelligence team conducted extensive user testing on our legacy VR simulations. The findings surfaced two clear recommendations for the design team: give learners a genuine sense that their actions matter, and provide opportunities to learn from mistakes. These became the core design constraints for the new design system of training simualtions.

Agency Drives Engagement

  • "I want to feel like my actions mattered." User testing research showed that the single strongest predictor of both user engagement and positive sentiment was whether users felt their actions genuinely impacted the simulation.
  • Agency fuels investment. The more control users have over their path, the more motivated they are to master the content and complete the experience.
  • Realism reinforces retention. Allowing users to experience the outcomes of both right and wrong decisions strengthens procedural and semantic knowledge through immersion.

Consequences turns errors into experience.

  • Mistakes that stick. Correcting errors in the moment transforms them into lived experiences, turning procedural and semantic knowledge into episodic memories. By engaging multiple memory systems, learning becomes more durable.
  • Emotion and consequence make learning memorable. High-stakes scenarios, where errors have visible outcomes, deepen engagement and long-term retention.
  • From passive to active learning. Real-time correction turns abstract rules into actionable skills, ensuring users know not just what to do, but why it matters..

Interactions Flows with Learning Outcomes

We designed the interactions path around two core, research-backed principles: user agency and immediate feedback. By giving learners meaningful choices and correcting mistakes in real time, every interaction became both an assessment and a learning opportunity. These patterns are flexible enough to adapt across a wide range of training content, yet structured enough to scale efficiently. Grounded in cognitive learning science, they provide a repeatable foundation that balances instructional effectiveness, user engagement, and production velocity.

Happy Path. Reinforce Correct Choices

  • Every interaction tied to a learning outcome gives the learner a clear choice.
  • Learners demonstrate understanding by making correct choices.
  • Immediate feedback confirms they’re on track without breaking immersion.
  • Positive reinforcement builds confidence and momentum.
  • Establishes a clear contrast to error-driven flows, highlighting success as part of the learning design.
  • Pedagogical purpose: reinforces knowledge through recognition and encourages continued engagement.
  • User Testing Insight: "Not everything was solved for me which I appreciate since I like to solve things by myself."
The learner makes the right decision, receives immediate positive feedback, and progresses smoothly through the interaction (1:03)

Correction Path. Learn Through Mistakes

  • Wrong choices prompt feedback without revealing the solution, encouraging reflection, retry, and active problem-solving.
  • Immediate coach feedback maintains immersion and shows that actions have real consequences.
  • Pedagogical purpose: Normalizes mistakes as part of learning, fostering resilience and problem-solving while ensuring learners actively build knowledge.
  • User Testing Insight: “I wish there was feedback for doing things wrong… anything to give feedback that your actions matter, instead of just motioning through a predetermined set of steps.”
The learner makes a mistake, experiences immediate correction, and re-engages with the task to find the right solution. (1:03)

Critical Error Path. Recover From Failure

  • Certain mistakes trigger a critical failure where the learner sees the real-world impact of a dangerous error (e.g., damage, injury, or job failure).
  • The immediate consequence creates a visceral moment that heightens awareness and makes the mistake memorable.
  • After the failure, the system resets and walks the learner step by step through the correct process in a guided tutorial.
  • This enforced repetition builds procedural memory and muscle memory, ensuring the correct actions stick.
  • Pedagogical purpose: Turns critical mistakes into powerful learning moments by pairing visceral awareness with immediate, guided remediation that reinforces the correct process.
  • User Testing Insight: "When you break a water pipe, there’s water coming out in the street so that’s a very nice touch as it shows what would happen if you make a mistake while doing this job."
The learner makes a critical error, experiences a failure state, then is guided step-by-step through the correct process to reinforce the right approach. (1:07)

Hint Path. Get Support When You Ask

  • When learners are unsure how to proceed, the system gradually offers help instead of leaving them stranded.
  • A hint button ensures the learner has a choice: they can ask for help when they’re ready, rather than being forced through looping reminders.
  • Support escalates in layers. First a clue and visual prompt, then a looping solution video if needed.
  • If learners still can’t complete the step, they can skip forward with context, ensuring the training continues.
  • Pedagogical purpose: Keeps learners engaged by preventing dead-ends, reducing frustration, and ensuring training can always move forward.
  • User Testing Insight: Dropout rates in legacy sims were closely tied to interactions learners could not get past. The Hint Path was designed to solve this problem.
The learner hesitates, receives layered hints, and is guided toward completing the step. (1:52)

Training Sim Prototypes

These prototype sims were built as proof-of-concepts to showcase the new design system in action, letting stakeholders directly experience decision-making and feedback cycles, and providing a tangible demonstration to align teams ahead of full production

Pre and Post Sim Prefab

We designed a modular start-and-finish flow that gives learners a clear on-ramp and a meaningful recap, while giving builders a reusable prefab for consistency. This flow solves legacy pain points, so learners begin prepared and end with a clear understanding of what they got wrong and how to improve.

Pre Sim Briefing

  • Preview and prepare. Briefing area previews tasks so learners know what to expect.
  • Tool familiarity. Tutorial table builds comfort with VR controls, reducing extraneous cognitive load in sim.
  • Learner control. Users choose when they're ready to begin.
  • Production consistency. Customizable prefab ensures consistent sim start experience and faster setup.
  • Legacy issue: Learners were dropped in cold, forced to learn VR controls and content simultaneously.
  • Benefit: Removes early frustration, lowers cognitive load, and ensures learners are confident and ready before tackling content.

Post Sim Assessment

  • Actionable feedback. New assessment system pinpoints exactly what was done incorrectly, supported by on-screen notes and clear voiceover.
  • Mastery through replay. Learners know how to improve, encouraging mastery and replay
  • Progress reinforcement. Passing scores unlock new content, rewarding persistence.
  • Legacy Issues: Scoring lacked clarity, with little guidance on how to improve.
  • Benefit: Learners leave knowing precisely what they got wrong, how to fix it, and how to improve their score the next time.

What I learned.

Systems accelerate learning. A well-structured design system helps simulations hit their learning outcomes faster by relying on pre-vetted workflows and recipes that are already proven to drive impact.

Consistency with flexibility. While content varies widely, aligning interaction flows to learning outcomes ensures a consistent learner experience that can still adapt to the needs of different scenarios and subject matter.

Leveraging VR’s strengths. Exploration, learner agency, and immediate correction of even critical errors harness what VR does best, turning mistakes and decision-making into safe, memorable, and impactful learning experiences

Teammates

Tools Used

Unity icon3DMax  iconPhotoshop iconAfterEffects icon