Rapid Prototyping Methods for VR & AR
Methods tested with Gabriela Madrid and Charles Paul Harris-White, and with the sponsorship of Valve.
Currently there are few prototyping tools for VR and AR, and most methods require a fair amount of coding and time to set up. With limited time, our team wanted to come up with a new way to rapidly test our concept and interactions without building a full product in a game editor.
Here are two of the methods we used to test and validate ideas, along with the strengths and weaknesses we found for each method. This was done as part of our team’s capstone study on how to improve the onboarding experience for new VR users and make menu interactions more intuitive.
BODYSTORMING
Our goals
Measure and gather data on what hand gestures and interactions would be intuitive for new users to interact with menu systems in VR.
Timeline
2 days to plan | 1 hour to create prototype | 1/2 day to tests
Method
We created a simple prototype using paper and ordinary office supplies to create 2D and 3D objects for our participants to interact with so we could test the difference in interactions. We came up with a scenario where we could observe them working with the menu system — building a pet house — which we believed would be easy for participants to understand and act upon.
Each participant was introduced to the testing procedure, and asked to imagine the scenario. We had participants walk through the scenario three times to see how controllers changed their interactions, and gather data on what each participant saw as the ideal interaction.
In the three rounds, participants demonstrated how they would interact with certain objects in (1)real life, (2) in VR using Valve’s controllers, and (3) we asked participants to think about their dream scenario of how they would complete the experience if there were no limitations. For the controller portion we had participants wear the controllers for the scenario, and were able to observe the differences with and without controllers.
Strengths of the method
- Little time required to create prototype
- Able to gather many ideas quickly
- Easier to build a connection between moderator and participant
- Able to see what they would do naturally with their hands and body
- Can compare natural vs. controller interactions
- Can have participants act out how they think something should work
Weaknesses of the method
- Not as immersive as it would be in VR
- Requires participants to use more imagination which could make them nervous and self-conscious
- Scenarios with complex feedback could be hard to prototype
MULTI-MEDIA PROTOTYPING
Our goal
Test and validate design solutions for interacting with menu systems in VR in a way that is interactive, immersive, and easy for us to observe.
Timeline
3 days to plan | 1 day to create prototype | 2 days to test
Methods
In order to demonstrate our research findings, we created a mixed-media prototype using PowerPoint, paper printouts, Valve’s controllers, and physical props to explore three of our VR features. We hypothesized that by presenting the user with a 2–3 variations of each feature, as well as allowing them the option to tell us how they would improve it, we could test and validate 8 menu interactions in a limited time.
For this test, we created an imaginary music game for participants to play so they would think they were testing the application and not think too hard about their interactions with the menu system.
For each participant, we had a short interview before the test to gather initial thoughts. We had them put on the controllers, and then we let them know that we were testing a game idea and to imagine that they were in VR. We projected the prototype and used Wizard-of-Oz techniques to provide feedback as the participant interacted with it. We assembled the prototype using designs from Sketch, photos, and motions brought together in PowerPoint.
We also provided props including actual musical instruments and items made of folded paper to create depth and observe 3D interactions.
Strengths of the method
- Much faster to prototype than high-fidelity VR/AR prototypes
- Doesn’t require expert development
- Easy to modify on the go
- Easy for moderators to observe and understand where and what the participants are struggling with
- Works well for testing interactions and concepts that should perform like real life
Weaknesses of the method
- Not all elements of a virtual environment can be simulated, so this test is not good for testing motion or haptic interactions
- 3D interactions are harder to test this way, so physical props are suggested to enhance the user experience
- Auditory and visual feedback can be hard to coordinate
- Unable to test extra-ordinary interactions that are possible in VR but not real life (such as flying, unusual physics, etc.)
- Requires an additional person to be the “Wizard of Oz” and manage the projection and the props
