How learning works

The guide to learning that actually sticks

👋 Welcome to Climate Drift: your cheat-sheet to climate. Each edition breaks down real solutions, hard numbers, and career moves for operators, founders, and investors who want impact. For more: Community | Accelerator | Open Climate Firesides | Deep Dives

---

Hey there! 👋
Skander here.

I don’t understand how learning works.

Or didn’t, until recently. For years, I treated learning like software updates: download the webinar, install the white paper, patch in the thought leadership. Then I’d wonder why nothing stuck past Tuesday.

The problem is the metaphor.

Your brain doesn’t work like a hard drive where information gets stored and retrieved. It works like a workbench where knowledge gets actively constructed, piece by piece, with severe limitations on throughput and a tendency to catch fire under stress.

Diana Hughes, who holds three patents on game-based learning systems that delivered 2-3x higher learning gains in RCTs and currently heads product for an AI teaching simulator, spent 12 years translating learning science into products that actually work.

Diana’s not here to tell you to take better notes. She’s here to rewire how you approach learning. Using the actual science of memory, retrieval, and cognitive load instead of the performance theater we mistake for learning.

🌊 Let’s dive in

---

How learning works

Introduction: The Climate Crisis is a Learning Crisis

Here’s something I’ve been thinking about a lot lately: we’re facing the most complex challenge humanity has ever tackled, and at its core, it’s about learning.

Think about it. Solving climate change requires us to teach the public new science, retrain millions of workers for green jobs, invent technologies that don’t exist yet, and help investors and policymakers understand entirely new frameworks. We need to learn how to build renewable energy systems at massive scale, how to redesign supply chains, how to change behaviors that have been ingrained for generations. The future of our planet literally depends on our ability to learn new things, fast.

Hi, I’m Diana! I’ve spent a lot of my career translating learning science into educational games, most recently games that teach kids to read and do math. But here’s the thing: the kids are going to have a hard time learning to read and do math if the planet is on fire.

As I’ve gotten deeper into the climate movement (shout out to Climate Drift Cohort #3), I’ve noticed something fascinating. The climate movement isn’t just about science and policy and technology - it’s fundamentally about learning. Every breakthrough we need, every behavior change we’re asking for, every new career path we’re creating, it all requires people to learn effectively, quickly, and in ways that actually stick.

But here’s the problem: most of us are terrible at learning. Not because we’re not smart enough, but because we’ve never been taught how our brains actually work.

So that’s what this guide is about. I want to give you some tools to be a better learner, because right now, we need to build a lot of new knowledge, fast. And the techniques I’m going to share aren’t just academic theories - they’re practical tools for the practical challenges you face everyday.

One request before we dive in: grab something to take notes with. Whether that’s pen and paper or a fresh document on your computer, actively restating what you’re learning in your own words is one of the most effective ways to reinforce new concepts. (There’s actually science behind why handwriting works especially well, which I’ll explain later.)

Okay, enough preamble. Let’s learn some stuff.

Your Brain Is Doing Its Best

TL;DR: Your brain is a magical pile of electrified meat. It is extremely powerful...but also kind of limited.

Let’s start with a fundamental misconception that trips up most learners: your brain is not a library. I know, I know - that’s what it feels like. You read something, it goes into your head, and later you pull it back out when you need it, right?

Wrong.

Your brain is actually more like a workbench. And understanding how this workbench operates is going to change everything about how you approach learning new climate concepts, from circular economy practices to the latest direct air capture technologies.

Enter your working memory. Despite the unhelpful name, working memory isn’t actually storage for things you remember. Instead, it’s where you make memories. Think of your working memory as a workbench where information comes in from the outside world through your senses - stuff you see, stuff you hear, and so on. This information arrives as soft, wet lumps of clay just sitting there on your workbench, ready to become useful.

Once they’re on your workbench, these squishy inputs can be shaped and hardened into bricks of knowledge, which can then be stored and used in your long-term memory. Got that so far? Info comes in, gets processed in your working memory, and then gets stored as knowledge in your long-term memory.

But here’s the catch: your workbench has severe limitations. You can only have a few pieces of information in working memory at one time (I’ve seen 3-4 as the limit, but there are others out there as well, so let’s go with “a few”), and you have about 10 seconds (again, ymmv) to process them or they fade away.

What happens when you hit those limits? Two things, and both of them will sound familiar if you’ve ever tried to absorb how flood risk modeling works or sat through a rapid-fire presentation on scope 3 emissions:

First, when you have too many inputs, they push each other off the bench before they can be processed.

Remember, your working memory can only hold a few lumps at one time. Anything beyond that is going to slide off and not get processed. This sort of thing can happen when you’re at the receiving end of a firehose of new information—you’ll process as much as you can as quickly as you can, but at some point, your workbench is full and anything new coming in isn’t going to get processed. You might kind of...feel it coming in through your senses, but you’ll discover later that you can’t actually remember any of it, because most of it didn’t make it across the workbench successfully. If you are a Driftie, you get to experience this weekly in the form of Skander’s deep dives :D

Second, if you leave those lumps sitting there too long, they dry out and blow away—often after about 10 seconds of inactivity.

So maybe you hear something important, you start to think about it, but then something else comes in...if you don’t get back to that first thought in time, it’ll be gone. Sound familiar? It’s called distraction.

For climate professionals, this explains a lot: why those 300-page IPCC reports feel overwhelming, why you can’t remember half of what was covered in that packed conference session, and why you lose the thread of virtual meetings when Slack notifications start popping up.

The good news? Now that you understand how your workbench operates, we can talk about how to optimize it. And I’ve got two practical tools for this.

Practical Tool #1: Dual-Coding

Remember when I said your working memory was like a single workbench? That’s not entirely true. Dual-coding theory says that we actually bring in information on two channels—verbal and visual/sensory. So words and more abstract concepts on one channel, and then straight sensory stuff like what you see, non-word noises you hear, textures, smells, etc. on the other.

Think of it as having two workbenches running in parallel. And here’s the performance gain: taking in the same information in two ways (like seeing a diagram and hearing someone explain it) is more likely to build strong knowledge than if you were just using one channel. It’s like you’ve encoded the same information in two different kinds of bricks.

This is why Kurzgesagt’s incredible explainer videos are so effective - they’re illustrating concepts as well as narrating them. Okay, there are many reasons why Kurzgesagt’s videos are amazing, but dual-coding is one of them.

There’s actually another channel, which I’ll call your physical/kinesthetic channel. This is when you use your body to help you encode information - think little kids clapping out the syllables in a word, or a dancer memorizing the steps in a routine.

But what if you’re stuck with single-channel input, like reading a report or listening to a presentation? In our modern multimedia world, we’re used to multi-coding, so when we’re only getting info on one channel, our brains start seeking input on the others whether we like it or not. And this is a recipe for distraction.

So what can you do? Deliberately fill your other channels with non-distracting input. Doodle while watching a video presentation. Listen to music while you write a report. I like to play match-3 games while I listen to Volts. You do you. The important thing is to give your brain the multi-channel input it craves without pulling attention away from what you’re trying to learn.

By the way, remember when I asked you to grab something to take notes with, preferably pencil and paper? Dual-coding is part of why that helps. You’re processing this article on the verbal channel and reinforcing it on the visual and kinesthetic channels by writing or drawing. Neat!

Practical Tool #2: Emotional Regulation (Don’t Light the Workbench on Fire)

We’ve talked about your working memory having limited throughput - 3-4-ish items at once, 10-ish seconds per item, right? Well, here’s something that can change that throughput dramatically: emotions.

Emotions act like modifiers on working memory. Generally speaking, emotions like fear, anxiety, and sadness will constrain your working memory.

Fear and anxiety in particular kind of light your workbench on fire. Like yeah, you can still make bricks on a flaming workbench, but it’s going to be a lot harder, and the bricks probably won’t be shaped quite right or hold up for as long.

This is the science behind test anxiety—when a student who knows the material bombs the test anyway because they were anxious about doing well. Constraining your working memory doesn’t just impact how you build new knowledge—it also affects how well you can pull out and use knowledge you’ve already built.

This works in reverse, too. Being in a good mood is great for your working memory! Who would you rather learn to cook from: Gordon Ramsay screaming about how raw your food is, or Roy Choi, perhaps the most delightful human alive? I know my pick, and it’s definitely the guy who taught me how to make a kimchi tuna melt, not the one who calls people “idiot sandwich.”

This is also why John Oliver’s approach works so well. Oliver specializes in delivering in-depth information on complex topics like carbon offsets, but sprinkled with humor and warmth instead of doom and panic. He’s keeping your workbench strong and functional while still covering serious material - a method I like to call “deep dives with dick jokes.” It’s hilarious, but it also works really, really well.

How do you fix this? Pay attention to your emotional state when you’re trying to learn. If you’re feeling anxious or overwhelmed, take a break. Do something that makes you feel good. Come back when your workbench isn’t on fire.

This is why I’m using so-simple-they’re-almost-dumb pictures and silly jokes throughout this guide. I want to keep your workbench sparkling and definitely not on fire. To be honest, I also do it because it’s more fun for me. Something I learned back in my game designer days: if the creator is having a good time, the audience probably will too.

Making Memories Is Hard

TL;DR: Knowledge is like a structure made of LEGO bricks, not like a filing cabinet full of unchanging documents

Alright, so we’ve talked about how your working memory is like a workbench where you turn mushy clay inputs into solid bricks of knowledge. But where do those bricks go? And more importantly, how do you organize them so you can actually find and use them later?

This brings us to schemas—the knowledge structures that live in your long-term memory. And here’s where I’m going to ask you to throw out another mental model that’s probably not serving you well.

Your long-term memory is not a filing cabinet. It’s not a library where knowledge sits on neat shelves, waiting to be retrieved. Instead, your brain stores knowledge more like...a restaurant.

I promise this will make sense. Eventually.

Think about everything you know about restaurants. You’ve got the basic pieces: food, eating, tables, menus, waiters, ordering, paying the bill. But these aren’t stored as separate, unconnected facts. Instead, they’re all linked together in a web of relationships. Hunger connects to food, which is connected to eating at a table, which is connected to a waiter bringing you the food, and so on. All these connected concepts form what learning scientists call a schema.

When you first learned about restaurants as a kid, you started with simple bricks: “someone brings me food,” “we eat at a table.” Over time, you added more bricks and connections: waiter, menu, ordering, tipping. Eventually, all these individual pieces “chunked” together into one big, dense concept: RESTAURANT. Now when you think “restaurant,” your brain doesn’t have to laboriously recall “menu + waiter + table.” It just pulls up the whole interconnected structure at once.

As you learn more, your chunks get bigger and more sophisticated. That basic “restaurant” schema can connect with “business,” which connects with “employment” and “market forces” and “capitalism.” Eventually you can understand more complex phenomena like why restaurant owners oppose minimum wage increases while servers support them.

What’s cool is, this takes the same cognitive effort as when you first learned “menu + waiter + table.” You’re still building schemas with the same limited working memory, but now your bricks are incredibly dense with meaning.

This principle is behind one of the most important insights in learning science: a novice is not a “little expert.” An expert’s schemas are fundamentally different from a novice’s because they’re built with those dense, interconnected bricks. Climate scientists don’t just know more facts than the general public—they see entirely different patterns and can use approaches that novices can’t even conceive of.

This is one way that climate communication can go sideways. Experts are operating at the level of “capitalism + market forces + policy incentives,” but novices are still at “menu + waiter + table.” So the experts use inscrutable jargon, reach conclusions based on data only insiders have, and generally just talk right past the general public…and then are frustrated when nobody feels as urgently as they do. The expert’s schemas are so dense, they can’t even see the tiny underlying bricks anymore. They’ve forgotten what it’s like to be a novice.

So how does this apply to your learning as a climate professional?

First, recognize that when you’re struggling with a new concept, you might not have the right schema to hang it on yet. If you’re trying to understand interconnection queues and it feels overwhelming, it might be because you’re missing some foundational chunks around grid infrastructure, transmission planning, or state vs. federal utility regulation.

Second, be strategic about schema building. Instead of jumping straight into advanced topics, make sure you’ve got the foundational schemas in place first.

Practical Tool: Metacognitive Monitoring

“Metacognitive monitoring” is a fancy term for “thinking about your own thinking.” In other words, as you learn, you need to pay attention to whether you are actually understanding what you are taking in, or if your shiny new schema probably has some big ol’ holes in it. Here’s how to do it:

Before you start learning something new: Ask yourself what you already know about this topic. What schemas do you have that might connect to this new information?

During your learning: Pause regularly and ask, “Can I explain this in my own words? How does this connect to what I learned earlier?” If you can’t make those connections, that’s a red flag that your schema isn’t forming properly.

After your learning session: Try to teach the concept to someone else (or at least explain it out loud to yourself). If you stumble, you’ve found gaps in your schema that need filling.

The goal isn’t just to memorize facts—it’s to build robust, interconnected knowledge structures that will actually be useful when you need them.

Using Memories Is Also Hard (retrieval practice)

TL;DR: You don’t just build a schema once. To keep schemas strong and useful, you need to regularly take them back to the workbench for maintenance.

Okay, you did it! You built a schema. You’re going to know that stuff forever now, right? Sorry. Even after you’ve built a solid schema, you’re not done. Your brain doesn’t just store that knowledge on a shelf where it sits pristine and ready for use. Instead, those knowledge bricks start getting a little squishy, and the blueprints for how they connect start to fade.

Remember how we talked about your brain being more like a workbench than a library? Well, here’s where that metaphor really pays off. Your brain doesn’t store schemas fully assembled. Instead, it stores the individual knowledge bricks and the blueprints for how they should be connected when you need them.

So every time you want to use your battery chemistry knowledge, your working memory has to pull those bricks out of storage and reassemble the schema according to the blueprints. But this isn’t just extra work, it serves a purpose: every time you do that reassembly, you have the chance to strengthen the connections, add new bricks, or connect this schema to others you’ve built.

But if you don’t use a schema for a while, those bricks get soft and the blueprints fade. Especially when the knowledge is new, the connections between ideas are fragile and need regular maintenance to stay strong.

This is why cramming for finals didn’t create lasting knowledge for you in high school. You might successfully reassemble your schema long enough to regurgitate information on a test, but then—poof—it’s gone. You never did the maintenance work needed to make those knowledge structures permanent.

The Science Behind Retrieval Practice

This brings us to one of the most counterintuitive things I have learned about learning: the best way to strengthen memory is to practice forgetting.

I know that sounds backwards, but it’s true. When you try to recall information that’s become a little fuzzy—when you have to work to reconstruct that schema from partially faded blueprints—you’re doing the cognitive equivalent of strength training. The effort required to rebuild strengthens all those connections.

This process has a name: retrieval practice. And it’s basically exactly what it sounds like. After you learn new material, you practice retrieving it from storage and using it.

Practical Tools for Retrieval Practice

Traditional approaches (that you probably learned in school):

• Flashcards - Classic for a reason, especially for foundational concepts

• Practice problems - Like the ones at the end of textbook chapters

• Teaching someone else - This is especially powerful because you have to both reassemble your schema AND think critically enough to help someone else build theirs

More creative approaches:

• Mind mapping - Draw out key concepts and their relationships (basically, draw your schema!)

• Concept diagrams - Especially good for understanding processes, like how low-carbon cement is made

• Rubber ducking - famously practiced by programmers, this is where you explain a problem out loud to an inanimate object (preferably an adorable, squeaky one) to surface the gaps in your own thinking

Modern digital tools that can help:

• The learning modes in ChatGPT, Claude, or Gemini - Just be sure to attach reliable information to have the LLM review with you so it doesn’t start making things up

• Spaced repetition apps - search for “spaced repetition app” on your search engine of choice and go to town

Oh, I haven’t explained spaced repetition yet? Okay, so:

Extra Credit: Spaced Repetition

If you really need to lock something into long-term memory, there’s an even more sophisticated approach called spaced repetition. This is basically a way of scheduling your retrieval practice to get maximum bang for your buck.

Here’s how most people study: they read material once, then cram right before they need it. This works temporarily, but the knowledge disappears quickly because the retrieval wasn’t effortful—those bricks and blueprints hadn’t had time to degrade enough to need real repair work.

The optimal approach looks different. You do your first retrieval practice session relatively soon after learning—while the knowledge is still fairly fresh but starting to fade. Then you wait a bit longer before the second session, longer still before the third, and so on. Each time, you’re letting your schema degrade just enough that reconstruction requires real effort, which strengthens it for next time.

This approach is especially valuable with material you just need to have memorized. Think doctors learning drug interactions, pilots drilling on emergency procedures, or chemists mastering the periodic table.

Building knowledge is like building muscle. If you want to get stronger, it’s not enough to lift weights every day - you need to lift progressively heavier weights, and you need to take breaks between workouts so your muscles can recover and repair. Why yes, I have been reading She’s a Beast, how did you know?

In a field as rapidly evolving as climate, this maintenance work isn’t optional. You need to continuously retrieve, strengthen, and update your knowledge so it stays useful and relevant.

Putting Things Into Other Brains

Alright, we’ve covered how you can optimize your own learning—dual-coding, emotional regulation, schema building, retrieval practice. But in climate work, it’s not enough just to learn. At some point, we also need to teach.

Whether you’re explaining renewable energy economics to investors, teaching communities about adaptation strategies, or briefing policymakers on the latest climate science, you’re constantly in the business of helping others build new schemas. Wouldn’t it be nice if you could speed things along by using bricks they already have? My friend, allow me to introduce you to:

Metaphors

TL;DR: Metaphors let you skip some of the hard work of schema building by borrowing bricks that already exist in someone’s brain.

Remember when I said that expert schemas are totally different from novice schemas? This creates a huge problem when you need to explain complex concepts to people who don’t share your knowledge base. But metaphors offer a clever workaround.

Here’s how it works: when you use a metaphor, you get to “borrow bricks” from schemas your listener already has. Instead of building new knowledge structures from scratch, you substitute familiar concepts that work in similar ways.

Consider how we talk about computers. When personal computers first emerged, people needed to understand entirely new concepts: file storage, data organization, software applications. How did we teach millions of people these abstract ideas? We borrowed from schemas they already had.

“Desktop” metaphor: Just like a physical desk where you organize papers and tools, your computer screen has a workspace where you arrange files and programs.

“Folders” metaphor: Digital files are organized in digital containers, just like paper documents in rainbow-bedecked Lisa Frank folders.

“The cloud” metaphor: flexible remote data storage as an amorphous, ever-changing vapor in the sky

These metaphors worked because they let people use existing schemas (office work, weather) to understand new concepts (data storage, remote computing). Nobody had to build computing schemas from scratch—they just had to map familiar concepts onto unfamiliar ones.

Climate communication is full of borrowed bricks too:

• “Carbon footprint” borrows from the physical trace you leave when walking

• “Greenhouse effect” borrows from the familiar experience of how greenhouses trap heat

• “Tipping point” borrows from the physical experience of something balanced on an edge - like a ball rolling over a hill’s crest - where a small push creates irreversible change

Each of these metaphors helps people understand complex atmospheric and economic processes by connecting them to things they already know. We’re not asking them to become meteorologists, we’re giving them just enough understanding to focus on what matters: this is a problem, here’s why, and here’s what we can do about it.

Metaphors also help with the emotional regulation piece. Instead of hitting someone with intimidating jargon right off the bat, you’re starting with knowledge they already have. This keeps them feeling confident instead of defensive or overwhelmed, which means their workbench stays functional instead of catching fire.

Practical Tools for Using Metaphors

Test your metaphors: Just because a metaphor makes sense to you doesn’t mean it works for your audience. Try explaining your concept to a few people and see where they get confused or where the metaphor breaks down.

Know when metaphors stop being useful: Every metaphor has limits. “Carbon footprint” helps people understand personal environmental impact, but it breaks down when you try to explain complex carbon accounting methodologies. Be ready to acknowledge when you need to move beyond the metaphor.

Layer your metaphors: Start with a simple metaphor to build basic understanding, then introduce more sophisticated concepts as your audience’s schemas develop. You might start with “greenhouse effect” and gradually work up to radiative forcing and feedback loops.

Metaphors are one of the most effective ways to bridge that gap between novice schemas and expert ones - not by dumbing down your expertise, but by connecting it to where your audience’s knowledge already lives.

Persuasion

This brings up something important that climate professionals often get wrong: teaching is not the same as persuading.

I hear this occasionally: someone will ask me, “Diana, how do I teach my climate denier relatives that they’re wrong?” And listen, I agree that they are very, very wrong. But that’s not teaching, that’s persuasion. The techniques we’ve been discussing—dual-coding, emotional regulation, schema building, metaphors—these are about helping people understand information. But persuasion? That’s about helping people change their beliefs or behaviors based on that information, and it requires a completely different set of skills.

As soon as your listener realizes that your “interesting fact” is actually a sales pitch, they’ll start to feel defensive and wary. Their workbench will catch fire, and all your careful schema building will go right out the window.

If you want to learn the art of compassionate persuasion, follow Renee Lertzman’s work. She’s far more expert than I am in the psychology of climate communication and behavior change. What I can tell you is how to help people understand information clearly - but what they do with that understanding is a whole other conversation. There’s a metaphor for this, I think. Something about horses…and water?

Anyway, the point is: persuasion and learning are two different things. Personally, I’m much better at the learning-and-teaching part than the persuasion part. That’s why I focus on giving you tools to understand information clearly rather than tools to change minds. But here’s what I do know: you can’t effectively persuade anyone if you don’t first understand the material yourself. And right now, there’s a lot of material to understand.

Conclusion: Bring Your Brain to the Fight

I started this guide by talking about how the climate crisis is fundamentally a learning crisis. I hope you now understand some of the science behind how learning actually happens. Your brain isn’t a library where you passively store data—it’s a workbench where you actively construct knowledge.

And right now, we need to build a lot of new knowledge, fast.

Every concept we covered—from optimizing your workbench with dual-coding and emotional regulation, to building robust schemas and maintaining them with retrieval practice, to using metaphors to help others understand complex ideas—these aren’t just academic theories. They’re practical tools for the most practical challenge humanity has ever faced: learning how to live sustainably on this planet.

Here’s what I want you to do next:

Take one technique from this guide and try it this week. Maybe it’s the metacognitive monitoring questions when you’re diving into a new climate topic. Maybe it’s using spaced repetition to really lock in some foundational knowledge. Maybe it’s being more intentional about finding the right metaphors when you’re explaining net zero emissions planning to your board.

Whatever you choose, pay attention to how it changes your learning process. Notice when your workbench is overloaded and needs a break. Notice when you’re building connections between new concepts and existing schemas. Notice when you’re successfully helping someone else understand something complex by meeting them where their knowledge already lives.

The climate movement needs people who can learn quickly, adapt to new information, and help others do the same. We need people who can absorb the latest climate science, master emerging technologies, navigate evolving policy landscapes, and translate all of that complexity for diverse audiences.

In other words, we need people with finely tuned workbenches.

You’re already taking a huge step by being part of the climate professional community. Now I challenge you to use these tools to build as much knowledge as you can, as effectively as you can. Because we need you, and all the brilliant things you’re going to learn.

The future of our planet depends on our ability to learn new things, fast. Let’s get to work.

---

Thanks and Further Reading

Want to know more about how learning happens? Have I got a book for you. Want to learn how to really make it stick? Have I got another book for you.

Gratitude and debt to these and many other authors, learning scientists, and practitioners whose theoretical work I have gratefully inherited, interpreted, and applied both for this article and my instructional design career more generally. Thank you for being so smart, and I hope you enjoyed my dumb pictures :D

---

Learning is better together. If you’re experimenting with these techniques or have your own strategies for making climate knowledge stick, I’d love to hear about it. Connect with me on LinkedIn or share what’s working for you in the comments.

---

If you like this essay, share it with the founder in your life:

---

And if you want to stay in the loop, don’t forget to hit subscribe: