Workshop Overview
A 3-hour hands-on workshop on creating AI-powered educational tools grounded in Cognitive Load Theory (Sweller 1988).
What You’ll Do
You’ll modify a working template to understand how to create personalized worked example generators:
- See the demo: Experience a complete application with 3 domains and 16 concepts
- Study the code: Understand how Pydantic structures AI outputs
- Modify and extend: Add concepts to a simplified starter template (1 domain, 3 concepts)
- Design applications: Plan how to apply this pattern to your own teaching
The Demo Application
The complete demo generates personalized worked examples in three domains:
- Programming (Python): For loops, list comprehensions, functions, etc.
- Health Sciences (Statistics): Correlation, t-tests, confidence intervals, etc.
- Agronomy: Yield prediction, NPK optimization, growing degree days, etc.
Learning Objectives
By the end of this workshop, you will:
- Understand the worked example effect and personalization principle from Cognitive Load Theory
- Modify a working template to add your own concepts (in your browser)
- Recognize the pattern for building educational AI tools with structured outputs
- Design extensions for your own teaching context
Get Started
Instructor
Dr. Andrew Ellis is a cognitive psychologist and researcher at the Virtual Academy at Bern University of Applied Sciences. He researches how artificial intelligence can be effectively used in educational systems.
Resources
- Workshop Materials: Start here
- Related Workshop: KI in der Lehre: Intermediate (Learning science foundations)
References
Sweller, John. 1988. “Cognitive Load During Problem Solving: Effects on Learning.” Cognitive Science 12 (2): 257–85. https://doi.org/10.1016/0364-0213(88)90023-7.