Productive Struggle by Design
Teaching in the Age of AI
Core Principle
As James Lang argues in Cheating Lessons (Lang 2013), “cheating is an inappropriate response to a learning environment that’s not working for the student.” Rather than focusing on student behavior, we must design learning environments that make productive struggle the most rewarding path.
Research on “desirable difficulties” (R. A. Bjork 1994; E. L. Bjork and Bjork 2011) shows that learning requires cognitive effort—when we remove struggle, we remove learning. Yet students are often not given incentives to engage with effortful processing. The goal is creating environments where students choose to engage in this effortful processing because they understand its necessity for deep learning.
Evidence-Based Environmental Design
1. Lower-Stakes Assessment (Lang 2013)
Replace high-stakes single assessments with frequent, lower-stakes evaluations:
2. Intrinsic Motivation Design (Lang 2013)
Build courses around meaningful problems rather than content coverage:
Coverage-Based: “Read Chapter 7 on climate change”
Problem-Based: “Analyze how climate patterns in your hometown have changed using local weather data”
Problems rooted in students’ environments may tap intrinsic motivation and therefore be less prone to AI generation.
3. Spaced Repetition (Cepeda et al. 2006)
Distribute practice across time rather than massing it:
Week 1: Introduce concept A
Week 3: Review concept A while learning B
Week 6: Apply concepts A & B to new context
Week 10: Connect A & B to concept CResearch shows spaced practice improves long-term retention by 200-300%.
4. Retrieval Practice (Roediger and Karpicke 2006)
Use testing as a learning tool, not just assessment:
5. Interleaved Practice (Rohrer 2012)
Mix different types of problems within assignments:
Instead of 20 algebra problems, then 20 geometry problems, mix them:
- Problems 1, 4, 7: Algebra
- Problems 2, 5, 8: Geometry
- Problems 3, 6, 9: Word problems requiring both
Interleaving improves discrimination between similar concepts and long-term retention.
6. Cognitive Load Management (Sweller 2023)
Design instruction to optimize working memory capacity while preserving productive struggle:
For Novices:
- Provide worked examples before independent practice
- Reduce extraneous cognitive load while maintaining intrinsic challenge
For Advanced Learners:
- Increase element interactivity gradually
- Encourage productive struggle through appropriately difficult tasks
- Focus on germane (learning-productive) cognitive load
Optimal Challenge Design
Create structured opportunities for effortful processing by finding the optimal difficulty level for learning:
- Too Easy: Students disengage; no learning occurs
- Too Hard: Students become overwhelmed; working memory overloads
- Just Right: Students experience productive struggle; learning occurs
Design Implementation:
- Start with student’s current capability level
- Add one new element of complexity
- Provide support for struggle, not elimination of it
- Allow adequate time for effortful processing
Subject-Specific Applications
Apply Interleaving:
- Mix problem types within practice sets
- Alternate between different mathematical procedures
- Include word problems requiring strategy selection
Use Worked Examples:
- Show step-by-step solutions before independent practice
- Gradually fade support as expertise develops
Implement Spaced Revision:
- Draft → Week gap → Revision → Week gap → Final edit
- Multiple low-stakes feedback loops
Design-Based Learning:
- Connect writing to real problems
- Authentic audiences beyond the instructor
Retrieval Practice Applications:
- Predict-observe-explain cycles with delayed feedback
- Concept mapping from memory before consulting notes
Desirable Difficulties:
- Generate hypotheses before receiving information
- Explain phenomena before learning the mechanism
- Struggle with “failed” experiments that reveal misconceptions
Cognitive Load Considerations:
- Worked examples for syntax, problem-solving for logic
- Pair programming as cognitive apprenticeship
- Debug others’ code to build pattern recognition
Implementation Guide
Reframing Difficulty for Students
Help students understand that cognitive effort is not a design flaw to be eliminated, but the very mechanism of learning:
Struggle Documentation Strategies
Building an Error-Positive Culture
Assessment Design:
Classroom Culture Shifts:
Structured Learning Activities
Pre-Learning Struggle (Generation Effect):
- Attempt problems before instruction
- Generate possible solutions to unfamiliar challenges
- Predict outcomes before demonstrations
- Create questions about upcoming material
During-Learning Struggle:
- Think-Pair-Share Confusion: Share what’s unclear, then work together
- Error Speed Dating: Rotate through common mistakes, discuss each
- Productive Failure Challenges: Deliberately tackle problems slightly above current level
- Explain-Before-You-Know: Attempt explanations before receiving information
Post-Learning Struggle:
- Transfer Challenges: Apply concepts to novel situations
- Error Diagnosis: Identify mistakes in sample work
- Peer Coaching: Help others through their productive struggles
Scaffolding Support Systems
The Struggle Support Ladder:
- Independent Attempt (5-10 minutes minimum)
- Peer Consultation (share struggles, not solutions)
- Resource Access (strategic hints, not full answers)
- Instructor Coaching (guide thinking, don’t solve)
- Collaborative Problem-Solving (work together on remaining challenges)
Strategic Hint System:
- Process Hints (not content): “What type of problem is this?”
- Strategy Suggestions: “What worked for similar problems?”
- Error Alerts: “Check your third step” (without giving the answer)
- Metacognitive Prompts: “What are you thinking right now?”
The Environmental Perspective
The goal is not to eliminate struggle, but to design learning environments where productive struggle becomes the most rewarding and supported path to achieving the outcomes we want for students—deep understanding, durable skills, and authentic expertise.
When environments align with how humans actually learn through effortful processing, students naturally choose deeper engagement because they experience the satisfaction of genuine capability development.
References
Reuse
Citation
@online{ellis2025,
author = {Ellis, Andrew},
title = {Productive {Struggle} by {Design}},
date = {2025-06-23},
url = {https://virtuelleakademie.github.io/denken-statt-delegieren/workshop/learning-environment/},
langid = {en}
}