Exploring spatial configuration and group familiarity when co-watching 360-degree stories in social virtual reality was posted to arXiv on 21 October 2026 (arXiv:2610.30353) by Portia Wang, Monique Santoso, Yujie Tao and Jeremy Bailenson, filed under cs.HC. The abs page carries no Comments field and no journal reference, so this is an unreviewed preprint.
Bailenson directs Stanford’s Virtual Human Interaction Lab, which has been running controlled studies on social presence since before most current headsets existed. The sample sizes here reflect that: N₁ = 109, N₂ = 116, N₃ = 88 — 313 participants across three experiments, which is unusually generous for immersive-media research.
The gap they are addressing
The framing is sharp:
While traditional cinematic viewing is often characterized as a social experience, engaging with immersive stories is predominantly researched as an individual activity.
That is correct and consequential. Film has been a room full of people for a century. VR narrative is studied, designed and evaluated as one person in a headset, and the research literature on presence is overwhelmingly about an individual’s relationship to a scene.
Social VR breaks the assumption, and in a specific way the authors identify: features built for ordinary social activities — virtual locomotion, spatial context, social context — become “meaningful levers for shaping the user experience” when the activity is watching a story together. Where people stand, how much room they have and whether they can wander are narrative variables, not just plumbing.
Three experiments, three levers
1. Spatial boundary type. Individual boundaries — each viewer in their own demarcated area — versus small communal boundaries shared by the group.
small communal spatial boundaries heightened social presence towards others compared to individual boundaries, but also increased mental demand
2. Size of the personal spatial boundary.
Reducing the size of the personal spatial boundary increased social presence towards others but also increased disorientation and frustration
3. Group familiarity — strangers versus people who know each other.
participants with higher group familiarity reported stronger attentional focus on the story, as well as higher spatial presence and social presence towards story characters, but also reported greater physical and perceptual demand
Read the three findings in a row and the structure is unmistakable. Every manipulation that made the experience more social also made it more effortful. The authors name it directly: “systematic tradeoffs between engagement and cognitive effort.”
What is genuinely interesting in result three
The first two findings are, broadly, VR proxemics behaving as proxemics usually do — crowd people and they feel each other’s presence and dislike it.
The third is stranger and more useful. Group familiarity increased social presence towards story characters and spatial presence — not just presence towards the other viewers. Being with people you know made the fiction more present.
That is not an obvious prediction. You could equally have argued that friends are a distraction from a narrative and strangers are socially inert and therefore easier to ignore. The finding points the other way: the people around you are part of how real the story feels, and knowing them makes the whole frame more credible.
It also came with higher physical and perceptual demand, which the paper does not resolve. Plausibly people with friends present move more, look around more and engage more actively — which is both why the story landed harder and why it cost more.
What a designer should do with this
The paper offers practical guidelines for designing shared immersive narrative experiences. The actionable shape, stated as decisions:
- Shared boundary or individual boundaries is a dial between intimacy and cognitive cost, and you should set it by narrative intent and by piece length. A ten-minute work can afford the mental demand of a communal space; forty minutes probably cannot.
- Do not shrink personal space to force intimacy. It works, and it buys the intimacy with disorientation and frustration — the two things most likely to make somebody take the headset off.
- Ask who is coming. A piece designed for groups of friends and a piece designed for a queue of strangers at a festival are different pieces. Most installation work currently assumes strangers by default without deciding to.
- Budget attention like a resource. If your spatial configuration is spending cognitive load, the narrative has less to work with. Dense dialogue and a crowded communal boundary are competing for the same thing.
Caveats worth stating
No peer review yet. All measures are self-reported — social presence, mental demand, disorientation and frustration are questionnaire constructs, and the paper describes the experiments as exploratory. There is no behavioural or physiological measure reported in the abstract to triangulate against. And “360-degree stories” is a narrower medium than VR generally: a fixed-viewpoint spherical video constrains locomotion in ways a fully explorable scene does not, so the proxemics findings may not transfer directly to a walkable piece.
What survives those caveats is the structural result, which is robust across three separate experiments and 313 people: in social VR, presence and effort move together. If a configuration seems to give you more presence for free, you have not measured what it cost.
Sources
- Exploring spatial configuration and group familiarity when co-watching 360-degree stories in social virtual reality — arXiv:2610.30353
- Stanford Virtual Human Interaction Lab