Introduction

This is the 3rd in a series of 5 articles on learning with AI. The series was inspired by a trend emerging across social and professional networks - post encouraging students to completely outsource their learning to AI tools, from generating summaries to creating flashcards, essentially bypassing the actual learning entirely. All to attract students by promising them less work and faster progression. In my exploration of learning in digital age, I've been writing about how human memory works and how AI tools can either hinder or enhance our learning process.

In the first article, Learning with AI - Introduction, we explored the three phases of learning: encoding, storage, and retrieval, described the forgetting curve, and discussed how overuse of AI tools can cause cognitive atrophy. While these tools offer access to information and learning support, they also present new challenges.

Our second article, Learning with AI - Encoding, focused on the encoding phase, particularly on elaboration techniques that help us understand and connect with new information.

👩‍🎓 Quick Recap - Encoding 
Encoding is what's happening right now as you read these words. You're taking in new information through your working memory, which is like your brain's active workspace. It's where you consciously process information, like when you're trying to understand a new concept or solve a problem.
Effective encoding happens when we actively engage with information by creating meaningful connections. This way, we remember it better than when we merely repeat it.

We also discovered that over-reliance on AI tools can lead students to skip crucial elaboration steps. When students let AI summarise, explain, or solve problems without engaging deeply with the material themselves, they miss out on the essential mental work that builds strong neural connections during the encoding phase.

Now, in this third article, we turn our attention to the storage phase of learning. We'll explore how information moves from working memory to long-term memory and why the timing of our study sessions matters more than their duration. The concept of spaced learning - distributing study sessions over time rather than cramming - has been proven highly effective for memory storage. We'll examine how AI tools can be leveraged to enhance this process rather than shortcut it, helping students develop more efficient and lasting learning strategies.

How Storage Works?

Think back to your last exam preparation. Did you find yourself cramming the night before, trying to force as much information as possible into your brain? You're not alone. Studies (1, 2) show that most students resort to cramming before exams, either by choice or because they've run out of options. While this approach might help you pass the test, it fails at helping you retain information for the long term.

To understand why, we need to look at how our brains store information - a process illustrated by Hermann Ebbinghaus's forgetting curve. This curve shows that without reinforcement, we forget roughly 70% of what we learn within 24 hours. 

But here's an interesting thing: our brains don't need long study sessions to create lasting memories. On the contrary, they thrive on repetition and spacing. When we revisit information at specific intervals, we're not just reminding ourselves of what we learned - we're actually strengthening the neural pathways that store this information. 

This is where spaced learning comes in - a method that aligns with how our brains naturally store information. Instead of trying to cram everything in at once (imagine trying to eat a week's worth of food in one sitting), spaced learning breaks study sessions into smaller, more frequent chunks spread out over time. Each session reinforces what you've learned and slows down the forgetting curve, making the information progressively easier to retain.

Image source

But why does this work so well? The answer lies in how our brains consolidate memories. When we first learn something new, our memory retention starts at 100%. Without review, this quickly drops off following a steep forgetting curve (shown in grey). However, spaced learning disrupts this natural forgetting process in a fascinating way.

Looking at the graph, we can see what happens when we review the material at regular intervals (marked by the dotted lines). Each review session bumps our memory retention back up to 100%, but more importantly, each subsequent forgetting curve (shown in blue) becomes less steep than the previous one. This means that after each review, we forget the information more slowly than before.

AI as Your Spaced Learning Assistant

ChatGPT has emerged as students' go-to AI tool for learning support, with other applications like Grammarly, Microsoft Copilot, Google Gemini, and Perplexity AI following (3). While many students use these tools primarily for content generation - writing essays, solving problems, or creating summaries - they also have the potential to support effective learning strategies and techniques.

Let's explore how these AI assistants can be particularly valuable in implementing and maintaining spaced learning practices, focusing on approaches that enhance rather than replace the learning process.

Here's how conversational AI assistants like ChatGPT or Claude can help enhance your study routine:

AI support role: Study Planning
Prompt: "I have an exam in [subject] on [date]. Here's my syllabus: [paste syllabus]. Can you help me create a spaced learning schedule for the next 3 weeks, assuming I can study 1 hour per day?"

AI support role: Breaking Down Topics
Prompt: "I need to study [topic] over the next two weeks. Can you break this topic into smaller subtopics that would take about 30 minutes each to review, organised in a logical learning sequence?"

AI support role: Interactive Review Sessions
Prompt: "I studied [topic] last week, focusing on [specific concepts]. Can you create 3-4 questions that help me review these concepts while making connections to [related topic] I'm studying today?"

AI support role: Building Connections
Prompt: "I've studied [topic A] and [topic B] separately. Can you help me explore how these topics connect and influence each other?"

Now that we understand how spacing out our study sessions strengthens information storage, let's look at how we can make each of those sessions even more effective. A technique called interleaving shows us that mixing up related topics within a study session, rather than tackling one topic at a time, can improve our learning outcomes.

Mixing it Up: Understanding Interleaving

Imagine your brain as a detective looking for patterns. When you study similar but different things side by side, rather than one topic at a time, your brain naturally starts noticing what makes each unique. This is the essence of interleaving - a learning technique where you mix up related topics or skills within a single study session instead of focusing on just one thing at a time.

While it might seem counterintuitive (wouldn't it be easier to master one thing before moving to the next?), research shows this "mixing it up" approach leads to better learning. A study by Kornell and Bjork (4) demonstrated this through art appreciation.

They designed an experiment where participants had to learn to identify different artists' painting styles. Here's what they did:

They showed participants 72 paintings and tested two different ways of learning:

1. The "Blocked" Method: Like studying one textbook chapter at a time, participants saw all paintings by one artist before moving on to the next artist's works.
2. The "Interleaved" Method: Like mixing up topics from different chapters, participants saw paintings from different artists in a jumbled order, constantly switching between artists.

You might think that studying one artist at a time (the blocked method) would make it easier to learn their style. However, the results were surprising: 78% of participants performed better when they learned using the interleaved method, where the paintings were mixed up. 

This research shows us something important about how our brains learn: when we have to constantly switch between different but related topics, we get better at noticing what makes each one unique. It's like our brain becomes more skilled at spotting the distinctive features that set things apart.

AI as Your Interleaving Assistant

Here's how AI tools can help with interleaving, along with specific prompts:

AI support role: Identifying Topics to Interleave
Prompt: "I'm studying [subject] based on this course book [paste course book]. Can you suggest which topics would work well together in an interleaved study session?"

AI support role: Creating Mixed Practice Sessions
Prompt: "Create a 45-minute study session mixing these three topics: [topics]. Include practice questions that help me switch between them."

AI support role: Finding Connections and Contrasts
Prompt: "What are the key similarities and differences between [topic A] and [topic B] that I should focus on when studying them together?"

AI support role: Designing Review Questions
Prompt: "Create a set of practice questions that alternate between [topic A], [topic B], and [topic C]."

Remember, the goal is to use AI as a support tool for creating effective interleaved practice, not as a replacement for the actual learning process. 

Key Takeaways

The storage phase of learning is critical for long-term knowledge retention. Here's what we've covered:

1. Spaced Learning Beats Cramming

• Regular, spaced reviews make memories more durable
• Each review session slows down the forgetting curve
• The same total study time yields better results when distributed

2. Interleaving Enhances Learning

• Mixing related topics in one session improves pattern recognition
• Helps spot differences and connections between concepts
• More effective than studying one topic at a time

3. AI Tools as Learning Partners

• Use AI to plan and structure spaced learning sessions
• Let AI help identify topics for effective interleaving
• Focus on active learning: do the work first, use AI for feedback

Remember: AI tools should support your learning process, not replace it. The most effective approach is to engage actively with the material and use AI to enhance your study strategies.

Next article in our series will explore the final phase of learning: retrieval.

Read more in:

1) Geller, J., Toftness, A. R., Armstrong, P. I., Carpenter, S. K., Manz, C. L., Coffman, C. R., & Lamm, M. H. (2018). Study strategies and beliefs about learning as a function of academic achievement and achievement goals. Memory, 26(5), 683–690. https://doi.org/10.1080/09658211.2017.1397175

2) Hartwig, M. K., & Dunlosky, J. (2012). Study strategies of college students: Are self-testing and scheduling related to achievement? Psychonomic Bulletin & Review, 19, 126–134. https://doi.org/10.3758/s13423-011-0181-y

3) https://universitybusiness.com/these-5-ai-tools-are-the-most-popular-among-students/

4) Kornell, N., & Bjork, R. A. (2008). Learning concepts and categories: Is spacing the “enemy of induction”? Psychological Science, 19(6), 585–592. https://doi.org/10.1111/j.1467-9280.2008.02127.x