The Curiosity Engine
In this webinar, Brent Proctor explores how AI, storytelling, and game-based learning can create more personalised and meaningful language learning experiences. Drawing on classroom practice and personal projects, he demonstrates how adaptive stories and interactive games can help learners engage with language through curiosity, choice, and real-world purpose. The session highlights practical ways AI can support learner motivation, personalise content, and transform language learning into a more immersive and engaging experience.
00:00:03
[Brent Proctor]
Hi, how are you? I'm good, thanks. I'm a little nervous, but we're good.
[Will Rixon]
Oh, you're going to be great. You're in a hobbit hole.
[Will Rixon]
Anyone in a hobbit hole is safe. Absolutely safe in a hobbit hole. Good stuff. So, Brent, you're Scottish educator based in Brazil. When did you move to Brazil?
[Brent Proctor]
Oh, about 12 years ago.
[Will Rixon]
12 years.
[Brent Proctor]
Yeah, maybe about 13 years now. But you'll hear a little bit about that in the speech.
[Will Rixon]
Yeah, great.
Okay. Okay, I won't dig too deep into that. Explore how stories, games, and AI can make learning more personal, playful, and get it. Your talk is really interesting, Brent, really close to my heart as well. I can't wait to hear it in full.
[Brent Proctor]
I love the way that you guys summarized it as well. It was the— I almost wanted to use your summary in the speech. And the talk itself.
[Will Rixon]
Okay. Okay.
Well, yeah, it was.
[Brent Proctor]
Yeah.
[Will Rixon]
Thank you very much, Brent. All right. Well, have a great talk and see you in about 15 minutes or so.
00:01:04
[Brent Proctor]
Thank you. Okay. Do a little bit. Oops. Someone else is sharing.
So share.
Okay. Is that sharing that?
[Will Rixon]
Almost. There we go. It's up. We're ready.
[Brent Proctor]
We're good.
[Will Rixon]
Okay.
[Brent Proctor]
I have my camera precariously balanced on my phone here. So what could go wrong, right? We'll be all right.
[Will Rixon]
That's great.
[Brent Proctor]
Okay, here we go.
00:01:39
So this all began with my kids. This is Melody. And Milo, and they were born in Brazil, and we'd lived there for about 10 years, so Brazil was basically the world that they knew. Then around 3 years ago, we moved back to Scotland to be closer to family, and as is often the case with family, things didn't quite work out the way we'd imagined, um, and also the weather in Scotland is awful. So eventually we came back to Brazil, um, and after we'd been back for a while, we noticed something with Melody.
She seemed unhappy. She'd get frustrated with little things. And naturally we did what parents do, we asked her loads of questions like, are you okay? Uh, do you want to talk about anything? But we got nothing.
00:02:38
Then one day, Yara my wife, this is Yara, she came home with a book, and I'll show you the actual book here.
The idea was that children could recognize their own feelings and situations through the characters. So instead of, tell me what's wrong, you could ask, how do you think she feels? What would you do in that situation?
And Melody started talking, like, at first about the characters, but then gradually about herself. And eventually we understood that during that year in Scotland, she lost a lot of her Portuguese. And she'd come back to Brazil, but suddenly she couldn't express herself at the same level as her peers.
00:03:33
And that was making friendships much more difficult.
Now, that, that really stayed with me, not least because watching your kids struggle to make friends is genuinely heartbreaking, but also because we'd managed to learn more from a story than we had from asking her directly.
And I just kept on wondering what else that story could do.
What if the child could actually change the story, like make a decision and see what happened because of it? Because that's life, right? It's curiosity, agency, and then consequence. And that became StorySteps. The way it works is The parent types in something the child is struggling with, a behavior or a situation, and StorySteps creates the first part of a story around that topic.
00:04:35
You read it together, then the child gets 3 choices. They pick one and the story continues from their decision. It doesn't really matter which one they choose. The important part comes at the end because now the parent and child can talk about what happened using the characters as a way in.
Now there was only one problem with all this. I had no idea how to build it. I'm an English teacher, I'm not a programmer, but AI had suddenly given me a way in. So I could describe what I wanted, review what it built, change it, break it, try again. And there was a lot of trial and error, like a lot of error.
00:05:25
But eventually I had StorySteps, and I remember looking at it and thinking, that's incredible. Look at what that story can do now.
I wonder what else it can do.
Now, around the same time, I'd started playing Dungeons and Dragons with Milo and Melody. If you've never played it, the basic idea is that someone creates an imaginary world, and the character you create has to make choices in that world, like whether or not to punch a goblin, right? And the dungeon master, or the storyteller, makes the world respond. Now originally it was intended to lure Milo and Melody to sleep, like a sort of interactive bedtime story. It didn't work because they got too involved with the characters.
00:06:19
They were in too deep. Uh, now I teach kids in Hong Kong from Brazil, which means I usually start work at 5 o'clock in the morning because of the time difference. And by bedtime, the prospect of being personally responsible for an entire fantasy universe was a bit daunting, right? So we got ChatGPT to be the dungeon master, and it was brilliant, but it was chaos because kids never do what you expect them to do. Milo and Melody would befriend things they were clearly supposed to fight.
00:07:00
Right, they developed their own side missions and they'd ask questions that nobody could have possibly anticipated, and the world just went with it. It adapted. And I loved that because nobody had to keep on dragging them back to the path if they became curious about something completely different. The world could follow them there. And inevitably, because I'm a teacher, I started wondering what happens if I bring that into class.
Now I teach a lot of children and parents ask me, can you get them to read more, right? Which sounds simple, but it isn't because you need something that they can read something that's useful for them to read, and rather inconveniently, something that they actually want to read too.
00:08:00
But now I could create something around the learner, and you can do this with your own students tomorrow. I described the Dungeons and Dragons setup, and then I typed, 'This learner is around A2 level.' keep the language accessible, and the world adapted to them. But then it struck me, if we'd been working on a particular grammar point or a specific vocabulary, why can't that appear naturally inside the story too? Not as an exercise, but in the way that the story is written.
So the learner meets the language while they're trying to follow the game, and because they care about the game, understanding it has a purpose.
00:08:57
That became the beginning of Fluency Quest, and it made me think much more seriously about why games are so good at getting information to stick. Because if you're a teacher, you've probably had this student, right? So you've been teaching irregular verbs for weeks. What's the past of go? Went.
Perfect. What's the past of see? Saw. Excellent. What did you do yesterday?
Yesterday I go to the park and see a duck.
Right?
And then that same child gets 1 hour of Minecraft at the weekend, 1 hour, and somehow they know everything. Obsidian, Nether portals, redstone mechanics, crafting systems, things I still don't understand despite Milo explaining them repeatedly.
00:09:57
And I'm thinking, how? You've studied English all week. You've played Minecraft for an hour. Why is this the information that's stuck? And I think part of the answer is that in Minecraft, knowledge has a job.
Now, the child wants something, and understanding the world changes what they're able to do about it. Minecraft doesn't stop and say, today's learning objective is to understand the properties of obsidian. The child learns about obsidian because suddenly obsidian matters.
Good games give you something to care about first and then knowledge becomes useful. The learning shouldn't interrupt the adventure. The adventure should create reasons for the learning to matter.
00:10:57
That's what I want English to feel like inside a learning game. I don't want a grammar exercise with a dragon meaning standing behind it. I want the language to have a reason to exist.
Now meanwhile, Yara, my wife, had been solving a much more practical problem. She's also an English teacher and she'd been taking full transcripts of her lessons and putting them into ChatGPT and asking it to help write student reports. And one day she said, could you just make me an app that does this? So I did, and that became Level Loop.
00:11:41
Now the basic idea is pretty simple, but so effective, right? Level Loop records the lesson, then creates a beautiful review, and over time those reviews build up. Into a map of that learner's progress and ability. So you can go back and ask Levelu things like, what does this learner keep struggling with? Like, what gets them talking?
What have they actually improved at? Which is incredible in itself, and I could do another speech entirely about that. But what did it mean for stories?
Well, I had a detailed record of what had actually happened in every lesson, and I thought, why don't I give all of that to Fluency Quest? Not just vocabulary or grammar. Now the whole lesson can enter the story.
00:12:50
Maybe they spent 10 minutes telling me about a day at the beach. That beach can appear in the story. Maybe they got a new puppy. The puppy can appear too. Maybe they talk about the hydraulic lift mechanics on a very specific model of a fire truck.
Put that in too. And suddenly the story isn't just adapting to the level of English, It's adapting to their lives.
And you can take that idea a long way. Imagine business English, man. You have a lesson on negotiation. Then afterwards, for homework, you actually enter a negotiation in the story. And you could make it episodic, a continuing storyline, recurring colleagues who remember you, relationships that develop, decisions that carry consequences into the next episode.
00:13:48
And the whole thing can operate within the rules, regulations of the real company that you work for, because we can just chuck that in too.
And now you've built a model of that organization, so why not use it for training? Here's a simulation of our company, take it home for a week, and make as many mistakes as you like. Come back next week and we'll review your results together.
And the more I think about that, the more the possibilities appear to be endless.
So what can a story do? Well, it can adapt its language to the learner.
It can make grammar and vocabulary useful because understanding them suddenly matters.
00:14:51
It can let somebody make a choice and see what happens because of it. It can put somebody inside a negotiation instead of just teaching them what to say. It can give them a safe place to make mistakes before they have to make those decisions for real. It can take something that happened in a lesson and carry it into a game. It can remember what the learner cares about.
It can adapt not just to their English but to their lives. And sometimes it can give a child a way to talk about something that they couldn't quite talk about before.
And every time I think I've worked out what a story can be, it seems to become useful for something else. So I, I think I'll leave you with the same question that got me into all of this in the first place: what else can it do?
00:15:56
That's the end, by the way.
Am I still, am I still going? Let's, I'll leave.
[Will Rixon]
You're there, Brent. Sorry, it took a while to load me in for some reason.
[Brent Proctor]
Ah, okay.
All good.
[Will Rixon]
Thank you so much.
[Brent Proctor]
Changed the ending a little bit, so you probably weren't expecting it. No, I was actually thinking—.
[Will Rixon]
I thought there was something else coming.
Great. Thank you so much, Brent. What a talk. What a talk.
[Brent Proctor]
Thank you.
Thanks.
[Will Rixon]
I love the— I love your embracing of technology and what actually people are doing in their private lives. You know what children are doing in their private lives. You know, my kids are doing it. He's playing.
00:16:48
You're so right about those games that they play. He could tell me a lot about Minecraft. And about mining, for example, because of that intrinsic motivation he's got of purpose in Minecraft. Yeah, I think it's really important. It sounds a little bit kind of, I don't know what it sounds like, but it's absolutely, it really chimes with me, Brent.
[Brent Proctor]
Thanks very much. Well, it means a lot to me as well. It's like a full spectrum of discovery happening, because which makes my work more interesting as well, to use it with kids and see how it performs. I think so.
[Will Rixon]
Yeah, yeah, yeah.
Well, thank you so much for your time, Bryn.
[Brent Proctor]
Yeah, thank you very much for having me. This is a, a highlight of my year.
[Will Rixon]
Oh, I'm so glad to hear it, Bryn. Thanks so much for coming.
[Brent Proctor]
It's been great to be here. Thanks a lot.