We Built a Game Where Kids Explore and Learn: A Teacher’s Notes on Playtesting and What We Need
The first time we put the game in front of kids, I was ready to explain the rules. I had a script. I had a list of learning objectives. I had a little timer set on my phone. The game was supposed to be about exploring a fictional ecosystem—kids would move through a forest path, discover small creatures, and answer questions about their habitats. But within five minutes, a boy named Miguel had climbed a virtual tree and was just sitting there, watching the leaves move.
He wasn’t solving anything. He wasn’t collecting anything. He had found a spot where the animation looped particularly well, and he was content to stay there. One of the other teachers leaned over and whispered, “Is that okay?” I didn’t know. I kept looking at my script. It didn’t have a box for “sitting in a tree.”
I teach second grade, and I’ve been working on this game in the evenings with a friend who codes. We’re not a company. We’re not a startup. We’re just two people who noticed that our students and our own kids often learn more when they’re allowed to wander than when they’re told exactly what to do. So we built a little world—nothing fancy—a flat green landscape with a few hidden areas. You can move around freely. There’s no wrong button. The game asks questions, but only if you want to answer them. We spent months making it, and then we gave it to twenty kids in our classroom to see what would happen.
What happened was that Miguel climbed a tree. And after a while, he came down, found a rock, pushed it over, and discovered a fact about moss. He read it twice. Then he went back to the tree.
I started to think about why that moment bugged me. I had planned for the game to be an adventure, with a clear path and a clear goal. But Miguel wasn’t looking for a goal. He was looking for a place to belong. When I watched the video later, I could see his shoulders relax. He was exploring the way kids explore a real playground—not to achieve something, but to feel something.
That’s when I realized the problem wasn’t with Miguel. It was with my assumptions about learning. In the classroom, I often set up stations: read this book, answer these questions, move to the next activity. It works for a lot of kids, and it works for me because I can measure it. But the moment the game let kids off the leash, they started making their own decisions. Some kids wandered. Some kids rushed to every corner. One girl just kept trying to see if she could break the game by walking into a lake. She couldn’t, but she was absolutely delighted we had built a lake.
I’ve learned that this kind of exploration is not the opposite of learning. It’s the foundation. Kids need to feel that an environment is safe before they’ll ask questions in it. For some kids, “safe” means knowing all the answers. For Miguel, “safe” meant knowing no one was watching him and that nothing was going to pop up and quiz him.
As we kept testing, we made small adjustments. The first version had a time limit on a puzzle, because I thought it would add urgency. But kids who took too long got anxious, and when they got anxious, they stopped noticing small details in the background. We took the timer out. It didn’t make the game harder or easier—it just made it more humane. We also added a “notebook” feature, a simple button that lets kids jot down something they noticed. We didn’t force it. We just put it there. To my surprise, kids used it more than I expected. One boy wrote, “lizard is sneaky.” That was it. That was the whole note. And that was learning.
There’s a practical lesson here for teachers and parents that has nothing to do with games. When we want kids to learn something, we often make the mistake of giving them a list. We say, “find three animals and write down their names.” But if we want kids to really pay attention, we can say, “find one thing that surprises you and tell me about it.” That’s a lower bar, and it invites curiosity. It also gives kids like Miguel room to sit in the tree first. They need that.
We’re at the point now where we need more eyes on this. The game works, mostly. Kids can navigate it, the text is readable, the sound doesn’t drive anyone crazy. But we don’t know how it works for kids who don’t like games. We don’t know how it works for kids who struggle with reading. We don’t know how it works for kids who would rather draw than play. We need teachers to try it in their classrooms and tell us what actually happens when they put it in front of a group.
If you’re a teacher and you’re curious, here’s what we’re asking: let your kids play for about twenty minutes. Don’t give them any instructions beyond how to get into the game. Then ask them two questions: “What did you notice?” and “What did you wonder about?” You can ask them out loud, on paper, or just watch and take notes. That’s it. We’re not looking for test scores or comparison data. We’re looking for the moments like Miguel’s tree—the moments where you can’t tell if the kid is learning or just messing around. Often it’s both, but you only see that if you’re paying attention instead of checking off objectives.
I’ve been a teacher for eight years, and I still get nervous when the lesson goes off the rails. But building this game has reminded me that some of the best learning happens when I stop trying to steer it. The kid who climbed the tree didn’t need me to tell him where to go. He needed me to trust that he was going somewhere.
We’re still working on the next version. We want to add more places to hide, more creatures that respond to how long you watch them, and maybe a few noises that only play if you’re still. I don’t know if it will be a game that teachers can use. I hope so. But even if it never becomes that, the testing process has already taught us something about how kids learn when we let them lead. Miguel is in third grade now. He doesn’t remember the game. But I do. And I’m starting to think I was the one being tested all along.
Please indicate: Thinking In Educating » We Built a Game Where Kids Explore and Learn: A Teacher’s Notes on Playtesting and What We Need