A touchscreen usually makes sense when visitors must choose, compare, or enter precise information. Gesture control may suit a short, visible, whole-body action when the space and audience support it. Neither input is automatically more engaging. Decide from the task a visitor needs to finish, then prototype it in the actual viewing and standing conditions.
Begin with the interaction, not the hardware
If the experience asks a visitor to select a product configuration, read details, and save a result, direct touch can make the choices explicit. If it invites a simple physical move that spectators can understand from the aisle, a gesture can become part of the experience. That is a design inference about task fit, not a measured rule that one technology always performs better.
For either option, write one sentence: “A first-time visitor can ___ in ___ seconds without staff instruction.” Then specify how they start, correct an error, and exit. A gesture system needs an understandable sensing area and confirmation that movement was recognised. A touchscreen needs controls large enough for the real setting and a route for visitors who cannot use touch. Test both in noisy light, with a queue behind the user, rather than judging a studio demo.
| Decision factor | Touchscreen question | Gesture-control question |
|---|---|---|
| Task precision | Can people accurately compare and select? | Can the action stay simple and unambiguous? |
| Public visibility | Can spectators see a result without seeing private data? | Can spectators tell which movement caused the result? |
| Physical effort | Can users reach and operate the screen? | Can users perform repeated motions comfortably? |
| Venue space | Can the screen and queue fit safely? | Is there a clear movement zone away from passing traffic? |
| Recovery | Can someone back up or restart? | Can someone opt in, opt out, and correct a misread? |
What the research does and does not tell us
Li and colleagues measured mid-air selection tasks using two interaction methods and developed a model of shoulder fatigue and recovery. Their results show why gesture height, duration, and rest periods belong in design review. They do not show that every gesture installation is tiring, nor that a brief waist-level action is equivalent to repeated above-shoulder selections. Li et al., ACM Transactions on Graphics, 2024.
Rodriguez and Marquardt asked 16 participants to propose opt-in and opt-out gestures for public displays, producing 430 suggested gestures. Participants generally preferred hand and arm movements, while acceptance was higher in an entertainment setting. The study explored expectations; it did not validate one universal gesture vocabulary for an event crowd. Its practical implication is to make participation voluntary and the start and stop states visible. Rodriguez and Marquardt, Interactive Surfaces and Spaces, 2017.
Read the two findings together: movement can be a powerful invitation, but a visitor still needs a low-effort, understandable way to join and leave. A spectacular interaction that requires a staff tutorial or a long arm-held pose may be a poor fit for an open aisle.
Prototype the choice in the venue context
Test first-time visitors with the task, not a question about which interface looks more exciting. Observe: can they discover how to start, finish, recover from a mistake, and leave? Ask whether the interaction was comfortable. Include visitors with different heights and mobility needs and provide a non-gesture route when a motion is required. For a camera-based system, show clearly when sensing starts and ends; decide what is captured, retained, and displayed before launch.
The UK Health and Safety Executive's guidance for managing crowds within a venue also matters: attractions and queues must not obstruct pedestrian routes. A gesture zone often needs more space than the display itself; a touchscreen can produce a tighter queue. Place the prototype and waiting area on a floor plan before approving the interface.
Archv's Formula 1 experience case study includes concepts for racing simulators. That kind of task calls for its own physical and operational test; it should not be treated as proof that one input method works best for every event. The buying decision is a brief for the visitor action, space, access, and staffing—not a hardware preference alone.