Over the last few years, virtual and augmented reality have become increasingly popular in education, but there is still a central, lingering question: How can students learn with this technology?
It’s still challenging for educators to find a way to get students to reflect on what they’re learning with a piece of hardware strapped to their heads. How do you get them to think more critically about the lesson without breaking their immersion in a virtual educational experience? Researchers at Northeastern University have found the answer in an unlikely place: escape rooms.
Immersive, interactive puzzle games that put players in a physical space and task them with finding clues to get out, escape rooms have become a global craze. They also could be the key to understanding how technology like VR and AI could be integrated into classrooms more effectively, according to a recent paper by Northeastern researchers.
The paper is published in Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems.
“Instead of going to learning scientists to talk about how people reflect and how you help people when they’re solving problems, we thought, ‘Well, escape rooms kind of do that already. Let’s talk to escape room facilitators [about] how they manage experience as well as problem-solving and what cues they look at, what they’re doing, what processes they have,'” says Casper Harteveld, a professor of game design at Northeastern.
Escape room facilitators, or game masters, are constantly observing, interacting and adapting to make the experience challenging yet accessible enough for players. Their approach should set a standard for AI development, says Harteveld.
The researchers on the project interviewed 13 escape room facilitators, or game masters, to learn their best practices and how they guide their players through an immersive experience like this. Through those conversations, they gained valuable insights into how immersive education can be better deployed, with the end goal of creating assistive “AI agents” that can help students in VR and AR learning environments.
Escape room game masters typically do two things when running players through a puzzle, Harteveld says. They observe their players through cameras that are hidden throughout the room, making sure players are progressing and finding clues. If players ever get stuck, GMs will then intervene by providing hints.
“We have a model of how GMs provide aid where they’re constantly observing what players are doing and then choosing whether or not to intervene based on their observations,” says Erica Kleinman, a postdoctoral researcher on the project. “They are constantly adapting everything they are doing based on how the players respond to the room, respond to them and respond to their aid.”
Some players might never want to receive help, while others might ask for more assistance. It’s the GM’s job to gauge when and how to get players to reflect on the puzzle in a way that doesn’t break immersion and that meets the players where they are.
“Escape rooms inherently already have a kind of reflective prompting built into them because you have the GM chiming in to help you,” Kleinman says. “It feels natural to have somebody say, ‘Hey, I noticed you’re stuck. Have you thought about what you did last time that you were stuck?'”
GMs have to constantly think about how well players are progressing, how they’re feeling as they progress and how to deploy aid when they might be struggling. One common method GMs use is what the researchers call an “incremental scaffolding of aid,” offering a vague hint and then gradually providing more specific hints if the players still need help.
“Personalization, adaptation, that’s where we want education to go, and escape room facilitators already do that,” Harteveld says.
Although there are lessons to be learned for educators, Harteveld says the end goal of this research is to design an AI model that can make VR and AR learning more productive for students.
By adapting the methods used by escape room GMs, they hope to create a tool that will assist with a student’s learning, not shortcut it.
“AI should augment our intelligence, and that requires an AI that scaffolds and helps and guides you along as opposed to ‘Give me the answer’ and you get the answer,” Harteveld says. “Basically, what we’re really developing is supportive reflection, because we want the AI to help you reflect on what you’re doing to improve and get you in the [right] direction.”
Moving forward, Harteveld says his team is creating an escape room that will allow them to test their AI model. The escape room is still in development, but Harteveld hopes their work will ultimately not only shape how immersive learning is used moving forward but how we think about our relationship with AI.
“Aid should make players feel like they came up with the solution themselves,” Harteveld says. “It’s really empowering the players, and this is something you can translate to AI. You still want the humans to do the work.”
More information:
Erica Kleinman et al, From Locked Rooms to Open Minds: Escape Room Best Practices to Enhance Reflection in Extended Reality Learning Environments, Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems (2025). DOI: 10.1145/3706598.3713811
Northeastern University
This story is republished courtesy of Northeastern Global News news.northeastern.edu.
Citation:
Escape rooms are fun, and they could also help make VR and AR effective tools for education and AI (2025, May 15)
retrieved 15 May 2025
from https://techxplore.com/news/2025-05-rooms-fun-vr-ar-effective.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.
