Guidelines for Designing AI Technologies to Support Adult Learning
This paper addresses the gap in AI educational technology design for adult learners by analyzing longitudinal deployment data from a US research institute to derive and validate 19 design guidelines, accompanied by a tool to facilitate their application in future system ideation.
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 teach a group of adults how to use a new, complex tool. Now, imagine that most of the "instruction manuals" and "design blueprints" you have were written for teaching 10-year-olds in a classroom. That's the problem this paper tackles.
The researchers found that while Artificial Intelligence (AI) is great at helping kids learn, it often misses the mark when it comes to adults. Adults aren't just "older kids"; they are busy professionals, parents, and workers with limited time, specific career goals, and a lot of life experience.
Here is a simple breakdown of what the paper does, using some everyday analogies.
The Problem: The "Child-Sized" Suit
Think of current AI education tools as suits tailored for children. They fit the classroom schedule, the short attention spans, and the dependency on teachers that kids have. But when you try to put that same suit on an adult who is juggling a job, a family, and a desire to learn a new skill on their own time, it doesn't fit. It's too tight in some places (too rigid) and too loose in others (doesn't address real-world goals).
The researchers looked at seven different AI tools developed by a national research institute (AI-ALOE) specifically for adults. These tools included things like:
- Virtual Tutors: Like a math coach that works 24/7.
- Smart Textbooks: Books that can chat with you and quiz you.
- Video Coaches: Videos that pause to explain why a step works.
- Social Agents: Digital matchmakers that help lonely online students find study buddies.
The Investigation: Listening to the Room
The team didn't just look at code; they listened to the people actually using these tools. They gathered thousands of comments from students, teachers, and the people who built the tools.
They used a method called "Affinity Diagramming." Imagine taking all those sticky notes with complaints and wishes (e.g., "I need help at 3 AM," "I don't trust what this bot says," "I want to know where my data is going") and sorting them into piles. They found patterns that kept showing up, no matter which tool or subject was being used.
The Solution: 19 Rules for Adult AI
From these patterns, they created 19 Design Guidelines. Think of these as a "Recipe for Adult-Friendly AI." They are grouped into three main categories, which the paper calls the "Community of Inquiry":
1. The "Brain" (Cognitive Presence)
- The Analogy: This is about how the tool helps you think and learn.
- The Rules:
- Be Flexible: Adults are busy. The tool must work on a phone, at 3 AM, or with spotty internet. It needs to fit into their life, not the other way around.
- Be Clear: Don't make the interface confusing. If an adult is tired from work, they don't want to figure out how to click a button.
- Be Honest: If the AI gives advice, it needs to explain why. Adults want to know the logic, not just get an answer.
- Let Them Drive: Adults are self-directed. The tool should give them choices, not just force a single path.
2. The "Teacher" (Teaching Presence)
- The Analogy: This is about how the tool fits into the actual class and helps the instructor.
- The Rules:
- Fit the Curriculum: The AI shouldn't try to be the teacher; it should support the teacher's specific style and the course goals.
- Connect the Dots: The tool should show the student how the lesson connects to their future job or career.
- Give Good Data: Teachers need dashboards that tell them, "Hey, 5 students are stuck on this concept," so they can help.
3. The "Community" (Social Presence)
- The Analogy: This is about not feeling lonely. Online learning can feel like shouting into a void.
- The Rules:
- Keep the Human Touch: AI should help the teacher, not replace them. Adults still crave human connection and empathy.
- Build a Tribe: The tool should help students find each other, so they don't feel like they are struggling alone.
- Teach Social Skills: It should help adults practice talking to people, not just talk to a robot.
The Reality Check: The "Heuristic Evaluation"
The researchers then took their 19 rules and tested them against the 7 AI tools they studied. It was like a "report card."
- What they did well: The tools were generally easy to use, looked good, and were grounded in good learning science. They were great at helping individuals learn facts.
- Where they struggled: The tools were weak on the "Social" side. They didn't do a great job of building community or helping students connect with each other. They also often hid how they handled user data (a big worry for adults).
- The Trade-off: No single tool got an "A" on everything. A tool designed to be a great math tutor wasn't designed to be a great social connector. The researchers realized that you might need a team of tools working together to cover all 19 bases, rather than one "super-tool" trying to do it all.
The Toolkit: The "Guideline Explorer"
Finally, the team built a digital tool called the Guideline Explorer.
- The Analogy: Imagine a map that connects the "Rules" (Guidelines) directly to the "Real Voices" (quotes from students and teachers).
- How it helps: If a designer is building a new AI and realizes, "My tool is confusing," they can open the Explorer, find the rule about "Clarity," and read exactly what real adults said they needed. It turns abstract rules into concrete ideas.
The Bottom Line
This paper argues that to build AI that actually helps adults learn, we can't just copy what works for kids. We have to design for the messy, busy, goal-oriented reality of adult life.
The 19 guidelines act as a compass. They remind designers that for adults, learning isn't just about memorizing facts; it's about fitting into a busy life, connecting with others, and feeling trusted and respected. The paper suggests that if we follow these rules, we can build AI that doesn't just teach, but truly supports the complex lives of adult learners.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.