The Metacampus initiative proposed VR as the meeting ground for research teams spread across institutions. That premise rests on an untested assumption: that existing collaboration platforms can carry real work, not just presence. Distributed research teams do not simply talk — they write, they annotate, they pull up references, they hand documents back and forth. Those are the tasks that decide whether a platform is a workspace or a novelty, and they were the ones nobody had systematically measured.

Participants

23 participants in the structured platform comparison, and 10 in the training study. Open-exploration sessions ran with intact research teams of at least three members, so that collaboration was observed between people who actually work together rather than between strangers assigned to a task.

Study Design

A mixed-methods programme of three linked studies combining quantitative, qualitative and observational data.

Study A was open exploration: research teams used VR naturally, with no prescribed task, to surface what they reached for and where they stalled. Study B was the structured comparison — 23 participants performing a fixed task set across four platforms: Horizon Workrooms, Spatial.io, Engage, and Mozilla Hubs. Tasks covered relocation, communication, writing, typing, and web browsing, chosen because they are the atomic actions underneath most collaborative research work. Study C compared CITI certification training delivered in VR against the same training on desktop, with 10 participants.

The hypothesis was not rejected: the platforms tested do not meet acceptable levels of usability for effective collaboration. Engage scored highest across UX metrics overall, while Horizon Workrooms returned the best SUS score among the commercially available options — a split worth noting, since it means the platform that felt best to use and the platform that scored best on the standard instrument were not the same one.

Two barriers dominated the qualitative data. Cognitive load was excessive — participants spent effort on operating the environment that should have gone to the work happening inside it. And simulation sickness became a reliable constraint past roughly 30 minutes, which is shorter than almost any real research meeting. A platform that degrades at the half-hour mark cannot host the thing it is being proposed for.

This work became the basis for Frustrations Associated with Using Commercially Available Virtual Reality Platforms for Productivity, presented at the IISE Annual Conference & Expo in 2023.

Testing four platforms against one task set was the decision that made the findings usable. Any single platform in isolation produces a list of complaints; the comparison showed which frustrations were product defects and which were properties of the medium — and the medium-level ones are the ones an institution needs to hear before committing to a rollout.

I also learned to be careful about what a good SUS score licenses you to say. Horizon Workrooms winning on SUS while Engage led the broader UX measures was a useful reminder that a single summary instrument compresses away exactly the detail a design team needs.