How to Make Anything Stick: Elaboration, Dual Coding, and the Two-Example Rule

Two students revise the same biology sentence: “Arteries have thick, elastic walls.” The first reads it four times, highlights it yellow, and moves on. It feels smooth. It feels learned. The second stops and asks a strange little question: “Wait - why would that be true?” She thinks, then answers herself: “Because blood leaves the heart in high-pressure bursts, so the walls have to stretch and bounce back without bursting.”

Guess who remembers it next week.

The first student mistook familiarity for understanding. The second did something small but powerful: she built the fact into a web of things she already knew. That single move - reasoning a fact out instead of memorizing it - is the heart of everything below.

In short: rereading feels productive and teaches you almost nothing. The techniques that actually make things stick all share one trait - they force your brain to do something with the information, not just look at it again.

Why this matters

You’ve probably spent hours “studying” that quietly evaporated. You reread the chapter, you highlighted the key lines, you nodded along - and a week later it was gone. That wasn’t a memory defect. It was a method defect.

Most of us were never taught how to study, so we default to the two laziest moves in the book: rereading and highlighting. They feel good precisely because they’re easy, and easy is the tell. The effort you avoid is the effort that builds the memory.

This post hands you three techniques that flip that around - elaboration, dual coding, and concrete examples - plus the memory-palace trick for the stuff that has no logic to it. None of them take special talent. They’re just things the top students already do, often without realizing it, and things you can copy starting tonight.

Elaboration: ask “why?” and build a web of roads

Elaboration means deepening a memory by connecting new information to what you already know - usually by asking why and how it’s true. Instead of storing a lonely fact, you tie it to a dozen nearby facts, so many mental paths lead back to it.

The most studied version has an intimidating name - elaborative interrogation - and a dead-simple habit behind it: when you meet a stated fact, ask yourself “Why does this make sense?” and generate the explanation yourself before checking.

Picture a fact as a single house in a huge city. If only one road reaches it, and that road is blocked, you’re lost. Elaboration builds extra roads. When recall time comes and one path fails, another still gets you there.

Its close cousin: self-explanation

Self-explanation is elaboration applied while you work through something step by step - you explain to yourself how each step follows and how it links to what you already know, instead of reading passively.

The classic evidence comes from Michelene Chi and colleagues, who watched students study worked physics examples. The high performers spontaneously talked through the examples far more - roughly 52 explanatory statements versus 18 for the weaker students - and they tied those explanations to the underlying physics. In a later study, students who were prompted to self-explain a biology text learned more than those who weren’t.

The lesson is quietly hopeful: the top students weren’t just born smarter. They were doing something different, and it’s something you can steal.

Why it works - and its honest limits

Elaboration works because of levels of processing: memory strength depends on how deeply you process something. In one famous test, people who judged a word’s meaning (deep processing) later recognized it far better - often about double - than people who only judged its typeface or whether it rhymed. Meaning sticks; surface fades.

But be honest about the ceiling. Elaboration leans on prior knowledge - if you know nothing about the heart, “why” questions have nothing to hook onto. And most of the supporting studies are short lab tasks, not year-long classrooms. Helpful, yes. Magic, no.

In short: turn each fact into a “why?” and answer it from what you already know - you’re building multiple roads back to the memory.

The generation effect: struggle first, then check

Here’s a rule that sounds almost too simple: producing an answer yourself beats reading the same answer. Psychologists call it the generation effect.

In the classic study, some people simply read word pairs like “rapid - fast.” Others got a rule and a cue and had to generate the second word themselves - “a synonym of rapid starting with f: ___.” The generators remembered far more. A later review of 86 studies pinned the average boost at a solid, dependable size.

Think of it like a muscle. Reading the answer is watching someone else lift the weight. Generating it - even straining and failing for a moment before you check - is your own muscle contracting. The effort itself is what strengthens the memory.

This is why elaboration and testing yourself are secretly the same family. When you ask “why do arteries have thick walls?” and generate the reason before peeking, you get the generation effect for free. The mistake beginners make is asking the “why” and then instantly reading the answer - that throws the whole benefit away. (It’s the same trap behind why rereading feels productive but teaches you nothing: comfort is the enemy of encoding.)

In short: always try to produce the answer yourself before you look. The mental effort of generating is what carves the memory in.

Dual coding: pair words with pictures

Dual coding is the idea that your mind stores information in two separate systems - a verbal one (words) and a visual one (mental images). Encode something in both, and you lay down two retrieval routes instead of one.

It’s like saving an important file in two different folders. If one path fails you at recall, the other still turns it up.

The evidence is striking. In one study people recalled more than twice as many pictures as words - the picture superiority effect. The clincher: when researchers told people to form a mental image of each word’s meaning, the picture advantage vanished. The words caught up, because people were now coding them visually too. That’s dual coding in action.

How to actually use it

Dual coding is not about decorating notes with clip art. The picture has to carry the meaning. Redraw a concept as a labeled diagram, a flowchart, a quick sketch - and crucially, put the visual right next to the words it explains, not on a separate page. You want both codes firing together.

Sketch a thick-walled tube with arrows showing a high-pressure surge and the wall stretching, and the words “arteries have thick walls” now have a second road home. If one route is blocked at recall, the other still reaches the memory.

Two honest cautions

First, don’t confuse dual coding with the learning styles myth - the popular but false belief that each person is a “visual learner” or “auditory learner” and should be taught only in their style. That idea has repeatedly failed testing. Dual coding makes the opposite claim: visuals help everyone by adding a second code, no “type” required.

Second, even why pictures win is debated. Some researchers argue the real reason isn’t a second code at all but distinctiveness - pictures simply stand out more. Tellingly, the picture superiority effect even shows up in people with aphantasia, who can’t form mental images at all. The theory is contested; the practical advice is rock solid.

In short: draw the idea and place the picture beside the words. Everyone remembers more with two codes - even if scientists still argue about why.

Concrete examples: from one story to the underlying pattern

Abstract ideas are slippery. Concrete examples - specific, familiar instances - give an abstract idea something to grip. But there’s a twist that trips up most learners, and it’s worth getting right.

The key study gave people a tricky puzzle: a doctor must destroy a tumor with radiation, but a beam strong enough to work would also destroy the healthy tissue it passes through. The solution is to aim several weak beams from different directions that converge on the tumor. Most people can’t crack it alone. But when learners first read two unrelated stories that shared the same “converge from many directions” structure - say, an army splitting up to attack a fortress from all sides - far more of them solved the tumor problem.

The magic ingredient was two examples, not one. With a single story, learners cling to its surface details. With two that share a deep structure, they can abstract the pattern - the skeleton underneath - and carry it to a brand-new problem. That ability to move a skill to a new situation is called transfer, and it’s the real goal of learning.

Concrete, but not too concrete

Here’s the twist. Examples that are too concrete and vivid can actually hurt transfer - learners get so absorbed in the surface details that they miss the abstract principle. Picture teaching fractions only with pizza slices: some kids end up thinking the rule is somehow about pizza.

So the recipe is: start concrete to build intuition, then fade toward the abstract structure, and vary the surface. Give an idea two or three different concrete examples, then ask the powerful question - “What do these all share?” That comparison is where durable, transferable understanding forms.

Approach	What the learner does	Result
One vivid example	Fixates on surface details	Poor transfer to new problems
Two or three varied examples	Compares them, extracts shared structure	Strong transfer
Abstract rule alone	Nothing concrete to grip	Feels hollow, forgotten fast

In short: give a new idea two or three varied concrete examples, then ask what they share. Comparison is what lets you transfer the idea to new situations.

A worked example: turning a dry fact into a sticky one

Let’s put it all together. Take one lifeless fact and rebuild it using everything above.

Before (dry fact, shallow processing):

“Arteries have thick, elastic walls.”

You could read that ten times and still blank on the exam. Now let’s elaborate, dual-code, and ground it.

After (elaborated, dual-coded, example-rich):

1. Elaborate - ask why, and generate the answer yourself first. “Why thick and elastic?” → “Because blood leaves the heart in high-pressure bursts. The walls must stretch to absorb each surge and spring back to keep blood moving. Thin walls would balloon and burst.” You’ve now reasoned the fact, not memorized it - and by generating the answer before checking, you grabbed the generation effect.

2. Dual-code - draw it. Sketch a thick-walled tube with arrows showing a high-pressure surge and the wall stretching, labeled “stretch + recoil = smooths the pulse.” Put that sketch right beside the sentence.

3. Concrete examples - give it two. Think of a garden hose that bulges when you kink it and pressure spikes; think of a springy rubber band that stretches and snaps back. Now ask: what do arteries, the hose, and the rubber band share? → they all handle pressure by stretching and recoiling. You’ve abstracted the principle.

The fact is now anchored by reasoning, a picture, and two familiar analogies. Three roads lead back to it instead of one.

The memory palace: a tool for arbitrary lists

Sometimes material has no logic to elaborate - a shopping list, the cranial nerves, foreign vocabulary. For pure arbitrary stuff, use mnemonics: deliberate tricks that impose vivid imagery onto material that has none.

The most powerful is the method of loci (Latin for “places”), better known as the memory palace. You mentally walk a familiar route - your home - and place a vivid image for each item at a specific spot. To recall, you take the walk again and “see” each image waiting there. It’s the technique behind Moonwalking with Einstein, where an ordinary journalist trains himself to memorize a shuffled deck of cards.

Say you need to remember three arteries - the aorta, the carotid, and the coronary. Walk your front hall:

• Front door: a giant OAR (aorta) blocking it, gushing water under huge pressure.
• Coat rack: a CARROT (carotid) wearing a scarf, pointing up toward the head and neck.
• Staircase: a golden CROWN (coronary) resting on the bottom step, circling a heart.

Recall is just walking the hall again and seeing each image. The sillier and more vivid the picture, the better it sticks.

Do mnemonics work? Yes - with limits

They work impressively for their narrow job. In one study, 40 days of method-of-loci training took memory-naïve adults from about 26 to 62 words recalled out of 72 - and their brain connectivity shifted toward the pattern seen in world-class memory athletes, an effect still visible four months later. For vocabulary, the keyword method has produced huge immediate gains - roughly 88% versus 28% for controls in one study.

But be clear-eyed. Mnemonics help you recall arbitrary facts - they do not build deep understanding or transfer. And their gains can be fragile: keyword benefits often fade faster over long delays than plain studying. So reserve mnemonics for genuinely arbitrary material, and always back them up with spaced repetition and review so they don’t decay.

In short: the memory palace is superb for memorizing arbitrary lists, but it aids recall of facts, not real understanding - and it still needs review to last.

Common misconceptions

A few comfortable habits quietly waste your study time. Here’s myth versus reality.

• Myth: “Dual coding means adding pictures to my notes.” Cute, irrelevant images just split your attention. Reality: the picture must carry the meaning - a relevant diagram, placed right next to the words.
• Myth: “Rereading and highlighting are studying.” They’re shallow processing that breeds false confidence; you mistake familiarity for learning. Reality: close the book and generate the idea from memory instead.
• Myth: “I’ll ask why, then read the answer.” That throws away the generation effect. Reality: struggle to answer first - even a wrong guess helps - then check.
• Myth: “One good example is enough.” One example traps you in surface details and blocks transfer. Reality: study two or three varied examples and ask what they share.
• Myth: “The more vivid the example, the better.” Overly concrete detail can hide the real principle. Reality: build intuition with the concrete, then fade to the abstract structure.
• Myth: “I’m a visual learner, so I only need pictures.” Learning styles are a debunked myth. Reality: visuals help everyone - and words plus pictures beat either alone.

How to use this

You don’t need to overhaul your whole routine. Pick these up one at a time this week.

1. Rewrite five notes as “why/how” questions. Take five facts, turn each into “Why would this be true?”, and answer from memory before checking your source.
2. Draw three concepts. Pick three ideas you’re studying and sketch each as a small labeled diagram right beside its written definition - words and picture on the same page.
3. Always study a second example. Whenever you meet a new principle, hunt down one more example of it, then ask yourself: what do these two share?
4. Generate before you peek. Make “guess first, check second” your default reflex - the strain is the point, not a sign you’re failing.
5. Build one memory palace. Choose a short arbitrary list (a grocery run, five vocab words) and place a vivid image for each along a familiar walk through your home. Take the walk again an hour later to test it.

Conclusion

If there’s one line to carry out of this, it’s this: memory is built by what you do to information, not by how many times you look at it. Asking why, drawing it, comparing two examples, generating before checking - every one of these forces your brain to build roads, and roads are what you drive back on when the exam or the meeting or the moment arrives.

But here’s the catch that trips up even people who use these techniques well: you can elaborate and dual-code a fact today and still lose it in a month if you never come back to it. The roads fade if nobody drives them. Which raises the question this whole toolkit leans on - when, exactly, should you review something so it sticks for good? That’s the quiet superpower waiting in how spaced practice beats cramming every time.