What is it about?

We studied how players react to a robot game master in a multi-party murder mystery game. We compared the robot physically handing out clues versus sending them to a digital screen. We found that a physical robot isn't always better. While comfortable players find physical interactions magical and highly engaging, players with high anxiety feel stressed by the robot invading their personal space. For these anxious users, receiving clues through a screen acts as a "social buffer". This shows that robots need to be adaptive, offering different modes depending on a user's comfort level.

Featured Image

Why is it important?

As robots become game masters, teachers, or helpers in hospitals, designers often assume physical interaction is the ultimate goal. Our research proves that a "one-size-fits-all" approach fails. The main lesson is that robots must be adaptive. If a user is nervous, the robot should switch to a digital mode (screens, text) to give them space. If they are comfortable, it can use physical mode to make the experience more magical. In short: don't force a handshake on someone who just wants a text message. The best robot knows when to get close and when to stay back.

Perspectives

Building our robot game master, POIROT, and watching players argue and collaborate was thrilling. The biggest surprise was seeing how a simple physical handover could completely overwhelm an anxious player, while others loved the sense of ritual. It made me realize that flexibility is the key to good robot design. We shouldn't just build robots that can approach people; we need to build robots with the social intelligence to know when a simple screen is exactly what the user needs to feel safe and engaged.

Wen Chen
Tsinghua University

Read the Original

This page is a summary of: POIROT: Investigating Direct Tangible vs. Digitally Mediated Interaction and Attitude Moderation in Multi-party Murder Mystery Games, March 2026, ACM (Association for Computing Machinery),
DOI: 10.1145/3757279.3788663.
You can read the full text:

Read

Resources

Contributors

The following have contributed to this page