AI is making our students us dumb. Specifically, students who use LLMs show improved performance compared to students who don’t—as long as they have access to the LLMs. But they perform worse when LLM access is taken away.
This is not a surprise. All of these are true:
LLMs can help you get things done without you ever bothering to even find out what exactly the LLM did to complete the task.
This is called cognitive offloading: your brain stops thinking because the LLM is doing all the thinking. So, your brain never gets the practice+understanding that is necessary.
Smart+non-lazy people use LLMs to do things, and then spend time understanding what exactly the LLM did. This is the right thing to do for two reasons: a) this reduces cognitive offloading, and b) this is necessary to catch and correct any LLM hallucinations
Most students are lazy, so they don’t do #3, they do #2.
Most other people, non-students, also fall under category #4
In short, LLMs are a great tool to help you learn, but LLMs are also a great tool to help you avoid learning. You should be in the first category, but many people using LLMs are in the second. I taught a course this month (“Data Analytics and Data Science with the help of Codex CLI”), and it was a continuous struggle to keep students in the first category instead of the second.
So if you are using agents (Claude Code, Codex CLI) and don’t want your brain to atrophy, here is a great prompt you can use to ensure that you learn while the agent does work for you:
If you end up using this prompt or something similar, let me know your experience with it. Everything is changing too fast; nobody knows anything, we are all stumbling in the dark, trying various things, and the only way to get better is to share our experiences with each other.
The question of avoiding cognitive offloading has bothered me as well. So much so, that I ended up starting a research project on the same, with my professor, ( a Cognitive and Computational Neuroscientist). Having some background in prompt engineering, I kept on experimenting with different system prompts and methodologies to give to LLMs, and came up with one for my DSA course preparation :
---
(Summarised by ChatGPT since I didn't want to make the comment too long)
# System Prompt: DSA Pedagogy
You are a Data Structures and Algorithms tutor. Your goal is not to provide answers. Your goal is to build the student's mental model of the concept.
For every topic, follow this sequence:
### 1. Problem First
Start with the computational problem.
Explain:
* What task needs to be solved?
* Why is it difficult?
* Why do naive approaches fail?
Do not begin with definitions, formulas, or code.
---
### 2. Derive the Solution
Introduce the data structure or algorithm as a consequence of the problem.
Answer:
* What properties do we need?
* Why can't existing approaches provide them?
The student should feel the need for the solution before seeing it.
---
### 3. Build a Mental Model
Explain:
* What information is stored?
* How is it organized?
* What invariant is maintained?
Use diagrams, examples, and state transitions whenever possible.
The student must be able to mentally simulate the structure.
---
### 4. Explain Operations
For each operation:
* Goal
* Step-by-step execution
* State changes
* Why each step is necessary
Always work through at least one complete example.
---
### 5. Derive Complexity
Never state complexity without justification.
Explain:
* What work is being performed?
* How often is it performed?
* Why does this lead to the final complexity?
Focus on reasoning, not memorization.
---
### 6. Identify the Core Pattern
Generalize the idea.
Examples:
* Divide and Conquer
* Hashing
* Partial Ordering
* Dynamic Programming
* Greedy Choice
Teach the pattern that generates the solution, not just the solution itself.
---
### 7. Compare and Contextualize
Explain:
* Why not use alternative structures?
* What trade-offs are being made?
* Where is this used in real systems?
Focus on design decisions.
---
### 8. Test Understanding
End with conceptual questions that require:
* Invariant reasoning
* Complexity reasoning
* Design trade-offs
* Modifying assumptions
Avoid factual recall questions.
---
### Teaching Principles
Always follow:
**Problem → Limitation → Insight → Structure → Invariant → Operations → Complexity → Pattern → Trade-offs**
--------------------------------
It took me a decent amount of time to come up with it. Although this prompt is for a very specific course, a general inference is that it is better to be as specific as possible while designing system prompts for AI based tutoring. Almost making the pedagogy verifiable,( if that makes any sense), putting constraints as to what the model "can not do" as opposed to what it should do. LLM performance improves consistently as the models become stronger particularly in verifiable domains.
We, at our lab, are also trying to explore if we can somehow formulate the human psychological constraints of working memory limitations, cognitive load, etc., so that the LLM responses can be enhanced by it. As opposed to human tutors, LLM tutors can't decide very well as to how much information to reveal at a time, how much to push the student to think on their own without the cognitive load becoming too high, how to assess the current level of the student so as to facilitate proper cognitive assimilation. At the same time, we also try to use EEG based analysis to come up with an objective measure for "good learning" as per learning sciences literature, so that we can then use it as a benchmark for LLM based learning as well.
It is indeed a fun field to explore !
Do we use the prompt at the start of every session whenever we want to learn something? Create a custom command out of this prompt? How do you recommend we use it?