Virtual reality technology is increasing in popularity with academics studying human behaviour, as it lets researchers put people into immersive scenarios and keep tight control over the experiment.

However, VR labs usually build their own custom software, using their own equipment, measurement tools and procedures. This makes it difficult for researchers to check or repeat studies from a different lab, something VR-related study relies on.

A team of researchers from around the world, including from UCL School of Management, have created a shared set of best practices for researchers using VR, making it easier to check and replicate studies, boosting the productivity and effectiveness of research using VR.

Co-author Dr Vivianna He, UCL School of Management said: “Currently, when original studies in psychology are repeated, the findings don’t hold up. VR can solve this because virtual environments can be recreated exactly, but this only works if researchers document and share their work properly. Allowing for easier replicating of studies would also lower costs related to building VR studies from scratch.”

The international team of researchers created a shared set of best practices in a regularly updated online checklist and repository for researchers, reviewers and ethics boards to use to evaluate a VR study.

They identified three areas where VR research needs common standards:

• Interoperability. The researchers recommend ensuring other labs can reuse experiments by using widely available 3D tools, keeping detailed records of digital assets used, being transparent about licensing and using an open technical standard called OpenXR, allowing a VR experience to run on different hardware
• Procedural standardisation. The researchers recommend a process to run VR studies using the same methods to avoid stark differences in results where different labs have used identical scenarios
• Data sharing. The team recommends following processes to ensure researchers have the required rights to share materials, using shared online repositories, pooling anonymised participant data and allowing controlled access when full sharing isn’t legally possible.

Vivianna added: “VR is a powerful lens into real human behaviour, but its value depends entirely on whether other researchers can trust and reproduce what they see. These protocols put transparency and shared standards first, so the technology stays accountable to good science rather than the other way round.

“My own projects on avatar-mediated team innovation processes and on how people build trust with AI avatars have shown me just how sensitive findings can be to implementation details that rarely make it into a methods section. That’s exactly the gap this PNAS article aims to tackle.”

The paper, Creating common virtual ground: Protocols to democratise open VR research was authored by a global group of academics and published in the Proceedings of the National Academy of Sciences.