Instructor-Created Custom GPTs as Pedagogical Partners Fostering Immersion in Online Higher Education: Two Case Studies
This paper presents two case studies demonstrating that instructor-created custom GPTs, when aligned with specific pedagogical goals and analyzed through the Immersive Learning Cube framework, effectively function as pedagogical partners to enhance student immersion in online higher education by simultaneously strengthening system, narrative, and agency dimensions.
Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine you are trying to learn a new skill, like writing a grant proposal or building software, but you are doing it entirely online. It's like trying to learn to swim in a pool where the water is deep, but the lifeguard is only available for 15 minutes a day, and the rest of the time, you're just floating alone. You might feel lost, distracted, or like you're just going through the motions without really "being there."
This paper is about two teachers who tried to fix that feeling of isolation by bringing in a special kind of AI helper (a custom GPT). But they didn't just use a generic chatbot; they built a specific "digital partner" designed to fit perfectly into their specific class.
Here is the simple breakdown of what they did, how it worked, and why it matters, using some everyday analogies.
The Big Idea: The "Immersive Learning Cube"
To understand if their students were truly engaged, the authors used a framework called the Immersive Learning Cube. Think of this cube as a 3D map of a great learning experience. To be fully "immersed" (deeply involved), a student needs three things:
- The System (The Environment): Is the water warm? Is the lifeguard there right now when you need them? (Responsiveness).
- The Narrative (The Story): Does the lesson feel like a coherent movie with a plot, or just a random collection of scenes? (Context and meaning).
- The Agency (The Control): Are you the driver of the car, or just a passenger? Do you feel like you have a say in what happens? (Autonomy).
The researchers wanted to see if their custom AI helpers could boost all three of these dimensions.
Case Study 1: The "Speedy Coach" for Grant Writers
The Setting: A super-fast, 5-week graduate course on how to write grant proposals (requests for money).
The Problem: It was so fast that students were drowning. They needed feedback instantly, but the human teacher couldn't be everywhere at once.
The AI Partner: A custom GPT named "GrantGPT" that acted like a 24/7 writing coach.
Boosting the System (The Lifeguard):
- Analogy: Imagine a coach who blows a whistle the second you make a mistake, rather than waiting until the end of the game.
- What happened: When students submitted a draft, the AI gave feedback in seconds. Students said, "It kept us working. We couldn't procrastinate because there was always something to fix right away." It kept the momentum going.
Boosting the Narrative (The Story):
- Analogy: Instead of just filling out a form, the AI helped them see their proposal as a story that was evolving.
- What happened: The AI didn't just say "fix this sentence." It said, "This part doesn't connect to that part; your story is breaking." It helped students see their proposal as one big, cohesive argument rather than a pile of disjointed tasks.
Boosting the Agency (The Driver):
- Analogy: The AI was a GPS, not the driver. It suggested routes, but the student had to choose which one to take.
- What happened: The teacher made it clear: "The AI gives options, but you decide." Students had to argue with the AI, accept some advice, and reject other advice. This made them feel like they were in control of their own learning, not just obeying a robot.
Case Study 2: The "Virtual Internship" for Software Engineers
The Setting: An undergraduate software engineering course in Portugal with 200 students.
The Problem: Online students often feel disconnected from the "real world" of coding. They also study at weird hours (late nights, weekends) when teachers are asleep.
The AI Partner: A custom GPT named "Catmming," who was framed as the company's "Intelligent Automatic Assistant" inside a fake software company simulation.
Boosting the System (The 24/7 Intern):
- Analogy: Imagine a senior engineer who never sleeps and is always at your desk to answer a quick question about code.
- What happened: Students used the AI late at night (11 PM, 2 AM) to get quick explanations. It reduced the "friction" of getting stuck and waiting for a human to wake up.
Boosting the Narrative (The Role-Play):
- Analogy: The whole class was a live-action role-playing game (RPG). The teacher was the CEO, the tutor was the CTO, and the students were trainees.
- What happened: The AI (Catmming) was part of the story. It didn't feel like a generic chatbot; it felt like a character in their "company." This kept the students feeling like they were part of a real team, even though they were at home.
Boosting the Agency (The Team Lead):
- Analogy: The AI was a consultant, not the boss.
- What happened: Students would ask the AI for advice on how to structure their code, but then they had to debate it with their teammates. They learned to say, "The AI suggested X, but I think Y is better because..." This forced them to think critically and regulate their own learning.
The Takeaway: AI as a "Pedagogical Partner"
The most important lesson from this paper is that AI shouldn't replace the teacher; it should be the teacher's sidekick.
- Before: Teachers were stuck grading tiny details (like spelling or basic code errors) and didn't have time to help with the big picture.
- After: The AI handled the "grind" (immediate feedback, basic questions), freeing up the human teacher to focus on the "magic" (storytelling, complex strategy, and emotional connection).
The Final Metaphor:
Think of the learning environment as a house.
- The AI is the smart home system: It turns on the lights instantly when you walk in (System), keeps the temperature perfect (Narrative consistency), and lets you control the settings (Agency).
- The Human Teacher is the Architect and Host: They designed the house, they welcome you, and they make sure the house feels like a home, not just a machine.
The paper concludes that when you combine a smart, custom-built AI with a thoughtful human teacher, students feel more "immersed," more in control, and more connected to their learning, even when they are miles away from the classroom.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.