Audience Participation Games: A Different Role for Viewers

Identify what a viewer can actually change

Audience participation games give viewers an input that can change the game being shown. In live game streaming, that creates a design question beyond whether somebody watches or comments: what operation does this spectator input perform, and whose experience is being evaluated?

Imagine a fictional clockwork arena in which an audience can move a shield protecting an automated figure. Sending a message that says “use the shield” and pressing a button that moves it are different contributions. Only the latter necessarily operates a game rule. This example describes a proposed interaction, without predicting that viewers will prefer it.

A 2024 study gave the audience influence over AI

Paliyawan and colleagues’ 2024 FightingICE system used cheer and jeer buttons to adjust AI fighters’ strength. It analyzed 74 questionnaires after 82 people participated across two classes. Everyone watched first, then encountered an interactive version; an adapted GUESS questionnaire compared experiences.

The participants supplying preferences were audience members. A software rule connecting their clicks to an automated opponent is evidence of an implemented input mechanism. It is not a measurement showing that a human contestant played better under observation. Keep that distinction when describing what this kind of interface makes possible.

A preference direction does not include every viewer

In both groups, directional enjoyment answers favored participation over watching, with reported binomial significance. Equal, neither and unclear responses were separate categories; the result did not mean every participant preferred interaction.

Imagine asking ten viewers about the fictional shield. Four choose the interactive version, one chooses watching and five give another answer. Interaction leads among the five directional responses, yet four of ten is not a majority of everybody. Those invented counts illustrate why the denominator belongs beside a preference claim; they are not additional study data.

The presentation and sequence limit the comparison

The paper’s Figure 5 group labels conflict with Table 4, preventing a confident mode-specific reading. Watching always came first, random assignment was not reported, and the exploratory analyses lacked a multiple-testing correction.

A viewer may learn the arena’s rules while watching and recognize more events in the next segment. Novelty could also change when a new control appears. This sequence cannot separate those possibilities from participation itself. The reporting discrepancy adds another reason to retain the common preference direction without ranking competitive and collaborative formats or assigning every outcome to one of them.

Describe the scope of influence before counting clicks

For the invented shield interface, write down who can move it, when a request becomes effective and how conflicting requests are handled. A viewer whose input is acknowledged may still be unable to tell which movement followed their request. Acknowledgment and collective influence therefore need separate records in this proposed arena.

Consider a single shield move selected by a vote versus unlimited movement from repeated button presses. Both offer viewer interaction, but they distribute influence differently. Click totals alone cannot explain which choice a person meant to make. These are original design comparisons, rather than further results from the FightingICE experiment.

Evaluate the spectator role in its own terms

A proposed playtest could preserve a viewing-only option and separately ask what viewers understood, whether their intended contribution occurred and which version they preferred. Record the input rules beside those answers. Someone who chooses to watch quietly has supplied a preference, rather than necessarily failed to engage with the game.

The fictional arena also needs a question about the spectacle: can observers still follow the contest while operating the shield? Answering that would require its own observations. Designing a spectator role means deciding what authority the audience receives and what evidence would show that people can understand and use it.

Keep exploring

FOLLOW THE SOURCE

Sources & further reading

Have a correction? Tell our editorial team.

✦ NICE. YOU LEARNED SOMETHING.