How do spatial boundaries and group familiarity shape co-watching 360-degree stories in VR?

Engagement gains arrive with cognitive costs

The paper asks how the spatial setup of a shared virtual viewing space, and how well group members know each other, change the experience of watching a 360-degree story together in social VR. Its answer is a set of tradeoffs rather than a single best configuration. Conditions that made co-watchers feel more present with one another, or made the story feel more absorbing, also tended to increase reported workload, disorientation, or frustration.

The evidence comes from three exploratory experiments with 109, 116, and 88 participants drawn from a university course. Each experiment manipulated a different factor: the type of spatial boundary around viewers, the size of the personal space allocated to each viewer, and group familiarity. Because the studies were exploratory and the familiarity manipulation check failed, the authors present the familiarity results as co-occurrence rather than a demonstrated causal direction.

Communal boxes beat individual seats on presence

The first experiment compared spatial boundary types, including a small communal box, general admission, and reserved seating with individual boundaries. Social presence toward other users differed by boundary type (p=.0050): the box condition scored higher than both general admission and reserved seating. Mental demand also differed (p=.0081), but in the opposite direction from what a presence-maximizing design would suggest: general admission and the box both produced higher mental demand than reserved seating.

The authors connect these patterns to behavior. Virtual movement differed by boundary type (p<.001), with general admission above reserved seating and the box above reserved seating. Interpersonal distance also differed (p<.001), with general admission greater than both reserved seating and the box, and reserved seating lower than the box. The share of the field of view covered by other users' avatars was highest in the box condition (p<.001), which the authors link to the higher social presence: when co-watchers occupy more of the visual field, users appear to feel more socially present with them.

A key caveat sits inside this comparison. In the reserved seating configuration, boundary type and the amount of available space were confounded, so the authors read the result as a difference between whole spatial configurations rather than as an isolated effect of either factor.

Smaller personal space raises presence and disorientation

The second experiment varied the size of each viewer's allocated personal space. Smaller personal space was associated with higher social presence toward other users (b=-0.070, SE=0.034, p=.041), and it also predicted greater disorientation (b=-1.07, SE=0.49, p=.030) and greater frustration (b=-0.11, SE=0.047, p=.019). Interpersonal distance increased with personal space size (b=0.14, SE=0.0072, p<.001), and the percentage of the field of view covered by other users decreased as personal space grew (b=-0.38, SE=0.040, p<.001).

The authors offer several explanations for why closeness cuts both ways. Users given less personal space had less agency to manage the social and spatial situation, and nearby avatars occluded more of the story. The story was rendered as a skybox, so the story world was spatially decoupled from the shared virtual room; nearby avatars that were poorly integrated with the story could therefore occupy more of the visual field and heighten perceptual conflict. The paper frames this as a possible mechanism, not a tested causal chain.

Familiar groups focus more but work harder

The third experiment examined group familiarity. Participants reporting higher familiarity also reported stronger attentional focus on the story (b=0.45, SE=0.22, p=.043), higher spatial presence (b=0.34, SE=0.12, p=.0048), and higher social presence toward story characters (b=0.24, SE=0.11, p=.027). The same participants reported greater physical demand (b=0.35, SE=0.14, p=.019) and perceptual demand (b=0.47, SE=0.23, p=.044).

Because the manipulation check failed, the authors treat these as associations. One interpretation they raise is that high-familiarity participants had fewer co-watchers in their field of view, which may have meant less effort spent monitoring others and more attention left for the story. The sample matters here: familiarity was low overall (M=1.73 on a 5-point scale), with no participant at the highest level and only one at the second-highest, which was later winsorized. The observed effects therefore describe variation within a narrow, low-familiarity range.

Attention toward others showed no measurable shift

Not every outcome followed the spatial manipulations. Attention from others and attention toward others showed no significant predictor, and the three-item attention-toward-others scale had low reliability (α=.22), so the authors examined two items individually and caution that null findings may reflect measurement noise rather than a true absence of effect. Oculomotor motion sickness also showed no significant predictor, and the number of users in the field of view did not differ by condition.

This distinction matters for reading the paper. The field-of-view coverage measure, which accounts for how much of the view avatars occupy, tracked social presence patterns, while simply counting avatars in view did not. The authors attribute that gap to overlapping avatars and interpersonal distance producing very different perceptual salience.

Design choices should follow the priority outcome

The practical reading is that no configuration wins on every measure. If the goal is social presence toward co-watchers, the small communal box and reduced personal space are the stronger options in these experiments, but both come with higher mental demand or disorientation and frustration. If the goal is lower mental demand, reserved seating with individual boundaries performed better on that measure, at the cost of lower social presence toward others.

Group familiarity adds a separate axis. Higher familiarity within this low-familiarity sample coincided with stronger story focus and presence toward characters, alongside greater physical and perceptual demand. The authors' own framing is that spatial configuration can partially offload social and spatial negotiation from individuals to the environment, which may modulate the mental effort of coordinating with others.

The paper's most decision-relevant point is therefore a comparison, not a recommendation: choose the spatial configuration by which outcome the shared viewing experience is meant to prioritize, and expect the engagement gain to be paid for in cognitive or perceptual effort. The authors also note that the studies aggregated behavior at the session level, so moment-to-moment dynamics and outcomes such as narrative comprehension and memory recall remain untested.

Sources used in this answer