Interactive exhibits are a staple of science and children’s museums around the world, and kids love them. Exhibits invite children to explore science concepts in a fun and playful way.
But do children really learn from them? Ideally, museum staff, parents or caregivers are on hand to help guide children through the exhibits and facilitate learning, but this is not always possible.
Researchers at Carnegie Mellon University’s Human-Computer Interaction Institute (HCII) have demonstrated a more effective way to support learning and increase engagement. They used artificial intelligence to create a new kind of hands-on, interactive exhibits that include an intelligent virtual assistant to interact with visitors.
When the researchers compared their smart exposure to a traditional exposure, they found that the smart exposure increased learning and time spent at the exposure.
“Artificial intelligence and computer vision have turned gaming into learning,” said Nesra Yannier, HCII faculty member and project leader, who called the results a “useful game.”
Earthquake tables are popular exhibits. In a typical example, children build towers and then watch them fall onto a vibrating table. Signs around the exhibit try to get kids thinking about science as they play, but it’s unclear how well they work or how often they’re even read.
Yannier led a team of researchers who built an AI-enhanced seismic table equipped with a camera, touchscreen, large display and an intelligent agent, NoRilla, which replaced the panels. NoRilla – a virtual gorilla – interacts with participants, taking them through different challenges and asking questions about why or not the towers fell along the way and helping them make scientific discoveries.
The team — CMU’s Yannier, Ken Koedinger and Scott Hudson; Kevin Crowley of the University of Pittsburgh; and Youngwook Do of the Georgia Institute of Technology – tested their smart earthquake exhibit at the Carnegie Science Center in Pittsburgh. Primary school-aged children attending a summer camp interacted with the smart or traditional exhibit and completed pre-tests and post-tests as well as surveys to assess what they learned and how how much they enjoyed the experience. The researchers also observed visitors interacting with the exhibit during normal hours.
Pre- and post-tests and surveys revealed that children learned much more from the AI-enhanced smart science exhibit compared to the traditional exhibit while having just as much fun. A surprising result was that although the children built more in the traditional exhibit, their building skills did not improve at all, as they mostly engaged in random adjustments rather than understanding the underlying concepts. . The AI-enhanced exhibit not only helped kids understand the [underlying] science concepts better but also transferred to better construction and engineering skills.
Their experiment at the Science Center also showed that people spent around six minutes on the smart exposure, four times the 90-second average of the traditional exposure.
“What’s particularly impressive to me is how the system engages kids in real science experimentation and thinking,” said HCII professor Koedinger. is required.”
Parents of children who viewed the exhibit said it was more interactive, guided and informative and offered two-way communication compared to other exhibits. They also said that “it uses inquiry-based learning, which is central to how children learn, but is also a play model, so it doesn’t appear to be a learning activity.”
“Our exhibit has automated guidance and support that makes hands-on physical experimentation a valuable learning experience,” Yannier said. “In museums, parents may not have the knowledge to help their children, and staff may not always be available. With AI and computer vision, we can bring that experience to more people. children from different backgrounds and on a larger scale.”
The team’s research began at the Children’s Museum of Pittsburgh, where they tested their smart exhibit design and made improvements based on feedback from people who interacted with it.
“This research will have lasting implications for future exposure experiments at the Science Center,” said Jason Brown, director Henry Buhl Jr. of the Carnegie Science Center. “Creating fun and inspiring exhibit experiences that support the learning and discovery of science, technology, engineering or math is what positions us as one of the most unique museums in the world. region.”
The team recently published their findings in the Journal of Learning Sciences. The Smart Science exhibit remains at the Carnegie Science Center as a long-term exhibit. It is also at the Children’s Museum in Atlanta and will soon be at the Please Touch Museum in Philadelphia and the Children’s Discovery Museum in San Jose, California.
“The Children’s Museum of Atlanta loves being a part of this research study. As we have watched the NoRilla in action, we are seeing high levels of ‘dwell time’ for both children and adults as they work to raise challenges through the combination of hands-on activities with computer challenges,” said Karen Kelly, Director of Exhibitions and Education at Museum of Atlanta. “We love that this experience aligns with our mission to stimulate every child’s imagination, sense of discovery and learning through the power of play.”
The CMU team is already working on creating other smart science exhibits using computer vision and AI to teach different science topics. Future plans include an exhibit with ramps and one with a scale.
Yannier pointed out that this technology will not only improve lessons in a museum, but could also help students learn in the classroom or at home.