Understand difficult concepts
Use AI as a study partner to break down confusing topics through dialogue, analogies, and examples tailored to how you learn.
← Back to New to AIWhy conversation works better than reading
Traditional studying: Read textbook → Highlight → Re-read → Hope it sticks
Dialogue-based learning: Ask questions → Get explanations → Ask follow-ups → Test understanding → Repeat until it clicks
Key insight
The Socratic method in action: The best way to learn with AI is through back-and-forth conversation. Don't just ask one question and accept the answer, keep questioning until it truly makes sense. This mirrors how great teachers work: they don't just lecture, they engage in dialogue.
Active vs. passive learning: Reading is passive. Dialogue is active. You're constantly retrieving information, testing it, and refining your understanding.
What makes dialogue effective
The customization advantage: Tell AI exactly what you do and don't understand. No judgment, no embarrassment about "basic" questions. A textbook assumes a certain level; AI adapts to YOUR level.
Example: "I understand what derivatives are conceptually, but I don't understand why the power rule works" is much more productive than just "explain derivatives."
Everyone learns differently: Some people need visual analogies. Others need step-by-step logic. Some need real-world examples. If the first explanation doesn't make sense, ask for a different one using a different approach.
The power of reframing: Often the concept itself isn't hard, you just need it explained from a different angle. AI can provide 5 different explanations in 5 minutes.
The Feynman technique: Try to explain the concept back to AI in your own words. AI will tell you what you got right and what you're still missing. This immediately reveals gaps in your understanding.
Immediate feedback loop: Unlike reading, where you might think you understand but really don't, dialogue forces you to actively demonstrate understanding.
A real learning conversation
Topic: DNA Replication (Biology)
Notice the pattern: The AI doesn't just explain, it asks questions that guide you to discover the answer yourself.
Let's start with the basics. What happens to the DNA double helix before replication can begin?
It unzips into two separate strands.
Good! Now, why does it need to unzip? What's the problem with trying to copy it while the two strands are still together?
Because... the bases are on the inside? So you need to expose them to copy them?
Exactly! The bases are hidden in the center of the helix, paired together. Once separated, what does each strand become?
A template for making a new strand?
Perfect! Each old strand serves as a template for building a new complementary strand. This is why DNA replication is called "semi-conservative." Based on what we just discussed, can you tell me what "semi-conservative" means?
Half is conserved? Like, each new DNA molecule has one old strand and one new strand?
Exactly right! You just derived the meaning from our conversation rather than memorizing a definition. That's deep learning.
What happened here
Guided discovery: Through a series of questions, you arrived at the concept of "semi-conservative replication" yourself rather than just being told. This creates stronger memory connections.
Building blocks approach: Each answer became the foundation for the next question. You built understanding progressively, from simple (unzipping) to complex (semi-conservative).
Immediate correction: When you showed uncertainty ("Because... the bases are on the inside?"), AI confirmed you were on the right track and helped you articulate it more clearly.
Request multiple analogies
Why multiple analogies work: Different analogies highlight different aspects of a concept. One might click when another doesn't. Seeing multiple angles reveals what's essential (appears in all analogies) versus superficial (specific to one analogy).
The multi-analogy prompt
When to use this: When a concept feels abstract or you can't visualize it. Analogies make abstract ideas concrete.
Why this prompt works
"As different as possible": Forces AI to explore different domains (physical objects, processes, social dynamics, etc.). This prevents getting three variations of the same analogy.
Breakdown request: By asking where each analogy fails, you learn the concept's boundaries. This is crucial, understanding where analogies break down prevents misconceptions.
Example: opportunity cost (economics)
Saturday night choice
You have one Saturday night. You can either go to a concert OR study for Monday's exam. If you choose the concert, the opportunity cost is the better grade you gave up by not studying.
Limited lunch money
You have $10 for lunch. Sandwich costs $8 (leaving $2). Salad costs $6 (leaving $4). Choosing the sandwich means giving up the salad PLUS the extra $2 you could have saved.
Phone storage space
You have 5GB left on your phone. A game needs 4GB, songs need 3GB. Download the game and you give up the songs AND 1GB of future space for photos.
What you learn
The common thread: Across all three analogies, opportunity cost isn't just "what you didn't choose", it's specifically the value of the BEST alternative use of that limited resource (time, money, space).
The key constraint: All three involve a limited resource that can only be used once. This is essential to opportunity cost, if resources were unlimited, there would be no opportunity cost.
Effective prompts for understanding
Explain at my level
When to use this: When you've tried to understand from a textbook or lecture but it's not clicking. Being specific about your background gets better explanations.
Why this prompt works
Context setting: By stating what you already know, AI can skip the basics and focus on your specific gap. This prevents wasting time on explanations that are too simple or too advanced.
"Don't skip steps": Textbooks often skip "obvious" steps that aren't obvious to beginners. This instruction forces complete explanations.
Socratic dialogue
When to use this: When you want deep understanding, not just surface knowledge. Best for concepts you'll need to apply or build on.
Why this prompt works
Active recall practice: By answering questions, you're constantly retrieving and testing your understanding. This is far more effective than passive reading.
Reveals hidden gaps: You might think you understand until you try to answer a question. The dialogue exposes what you're missing.
"Until I can explain it correctly": Sets a clear success criterion. You're not done when AI has explained it, you're done when YOU can explain it.
The "Explain It Back" Test
When to use this: After AI has explained something. This is the crucial verification step that most students skip.
Why this prompt works
The Feynman technique: If you can't explain it simply, you don't understand it. This forces you to articulate your understanding, which reveals gaps.
Diagnostic precision: AI can pinpoint exactly where your understanding breaks down. Is it the basic concept? The application? A specific detail? This tells you where to focus review.
Prevents illusion of understanding: Reading an explanation and nodding along feels like understanding. Explaining it back proves whether you actually get it.
Do's and don'ts for learning with AI
Vague: "I don't understand calculus"
Specific: "I understand that derivatives measure rate of change, but I don't understand why the power rule (bringing down the exponent) works mathematically"
Why specificity matters: The vague version could lead AI to explain all of calculus from scratch. The specific version targets exactly what you need, saving time and giving more useful explanations.
Why both matter: Understanding what something IS is only half the picture. Understanding what it ISN'T prevents common errors.
Example: "Give me three examples of opportunity cost and three examples of things that seem like opportunity cost but aren't." The counter-examples clarify boundaries.
Don't stop at the first explanation: Ask "Can you explain that differently?", "Can you give another example?", "What's a common misconception about this?"
The goal: Multiple angles on the same concept. Each explanation adds a piece to your understanding.
Wrong approach: "Solve this problem for me: [paste problem]"
Right approach: "I'm stuck on this problem. I think I need to use [concept] but I'm not sure how. Can you explain the approach without solving it for me?"
Why this matters: Getting answers helps you finish homework. Understanding concepts helps you pass exams and actually learn.