Why Your Best Ideas Keep Dying in the Feed — And the Cognitive Trap Behind It

Reading time: ~12 minutes

In 1990, a Stanford PhD student named Elizabeth Newton ran an experiment that should have permanently humbled every smart person on the internet.

She had volunteers tap out the rhythm of well-known songs on a table — Happy Birthday, The Star-Spangled Banner, songs everyone knows. Their job was simple: tap the rhythm. Their partners’ job was simpler: guess the song.

The tappers, hearing the song clearly in their own heads, predicted their partners would guess correctly about half the time. The actual success rate was 2.5%.

The song that was so loud and obvious in the tapper’s head was, on the table, just a series of nonsense thuds. The information had not crossed.

This finding — eventually labeled the Curse of Knowledge by economists Colin Camerer, George Loewenstein, and Martin Weber, and later popularized by Chip and Dan Heath in Made to Stick — explains more about why intelligent people struggle to be heard than any other single concept in cognitive science. If you’ve ever written something you were proud of, sent it into the world, and watched it die in silence, you have been the tapper. You hear the melody. Your audience hears thuds.

You hear the melody. Your audience hears thuds. The gap between them is not a failure of intelligence — it’s a failure of translation.

This article is a long-form treatment of why this happens, why it gets worse as you become more expert, and what specifically to do about it. We’ll cover the cognitive science, the structural mistakes intelligent writers make, and a complete framework for translating sophisticated ideas into formats that actually spread.

Table of Contents

1. The Cognitive Science of Being Misunderstood
2. Why Intelligence Makes the Problem Worse
3. The Clarity Tax
4. Why “Basic” Advice Outperforms Brilliance
5. The One-Variable Rule
6. The 80% Rule
7. The Dinner Party Test
8. A Framework for Translating Hard Ideas
9. Common Objections (and Why They’re Traps)
10. Key Takeaways
11. Frequently Asked Questions
12. Further Reading

1. The Cognitive Science of Being Misunderstood

The Newton tapping experiment is famous, but the underlying mechanism is worth examining carefully — because once you understand why the curse operates, you understand why willpower alone cannot defeat it.

What Happens in the Tapper’s Brain

When you tap Happy Birthday on a table, you don’t experience a series of physical impacts. You experience the song. Your brain is silently filling in the melody, the lyrics, the emotional associations, the mental image of a cake. The tapping is just the visible sliver of a much richer internal experience.

This filling-in is automatic and uncontrollable. You cannot, by force of will, hear the taps as a stranger hears them. The knowledge has fused with the perception. Once you know the song, you cannot un-know it.

The same phenomenon happens whenever you write or speak about a topic you understand deeply. Your sentences trigger, in your mind, an entire ecosystem of context — examples, exceptions, history, related concepts, the reason this idea matters. The reader sees only the sentences. The ecosystem stays in your head.

The Receiver’s Problem

Your audience is not failing to listen. They are simply receiving a smaller signal than you’re sending. You wrote ten things; they received three. You implied seven; they noticed two.

This is not a defect in their attention — it is the structural reality of communication: the receiver always works with less than the sender intended. Add normal life — distractions, fatigue, competing inputs — and the receiver’s bandwidth shrinks further.

By the time your idea reaches them, it has been compressed, mis-decoded, and partially ignored. What survives is not what you said. It’s what they were able to extract.

Why “Just Write Better” Doesn’t Help

The intuitive response to this problem is to write more carefully. More precision. More qualifications. More definitions. More footnotes.

This makes the problem worse. Each additional precision adds load to the reader. Each qualification slows them down. Each footnote demands a context switch. The smart writer’s instinct — to bullet-proof their ideas against misinterpretation — produces texts that are mathematically harder to read, and therefore more likely to be misunderstood, abandoned, or ignored.

The fix is not better writing. The fix is a different theory of what writing is for. This is the same dynamic explored in The Analysis Paralysis Trap — intelligence, when misdirected, actively works against the outcome you’re trying to achieve.

2. Why Intelligence Makes the Problem Worse

There’s a particular failure mode that afflicts highly intelligent people, and once you see it you cannot unsee it.

The Preference for Nuance Over Resonance

Smart people don’t just suffer from the Curse of Knowledge. They actively prefer nuance over resonance. They would rather be 100% accurate and unread than 90% accurate and shared.

Deep down, they feel — ethically — that simplifying an idea is a betrayal of the truth. The intelligent writer believes that to leave out the edge cases is to mislead the reader. To skip the qualifications is to lie. To pick a clean example over a messy one is to falsify the data.

That position is admirable in a graduate seminar. It is suicidal on the open internet.

Because the open internet does not reward completeness. It rewards transmission. An idea that crosses successfully from one mind to another — even in slightly compressed form — has done more good in the world than a perfect idea that never crosses at all.

The Performance Trap

There’s a deeper version of this problem. Many intelligent writers are not actually writing to communicate. They are writing to perform intelligence.

The audience is not the reader; the audience is some imagined panel of credentialed peers who will judge whether the writing is sufficiently sophisticated. You can spot this writing easily. It uses jargon when plain language would do. It piles on caveats no normal reader needs. It refuses to use simple analogies because simple analogies feel undignified.

Above all, it performs depth by adding complexity rather than achieving depth by removing unnecessary parts. This is writing that protects the writer’s identity at the expense of the reader’s understanding. It is, in a real sense, selfish.

This connects directly to a pattern documented in Why Your Smartest Employees Make the Riskiest Decisions — the same cognitive overconfidence that makes experts write opaquely also makes them act outside their remit. The underlying mechanism is identical: expertise breeds blind spots.

The Audience Mismatch

Most public writing is judged by people who have spent zero time on your topic. They didn’t take the prerequisite course. They haven’t read the foundational paper. They don’t know the history of the debate. They are simply trying to extract one useful thing from your piece in the seven minutes they’ve allocated.

Writing for the imagined panel of peers means losing the actual audience of curious readers. You cannot serve both. Choose deliberately.

3. The Clarity Tax

Think of every sentence in a piece of writing as charging the reader a small fee. The total fee, summed across the piece, is the Clarity Tax.

What Counts as a Tax

• Sub-clauses are taxes.
• Parentheticals are taxes.
• “However” and “although” and “that said” are taxes.
• Undefined acronyms are taxes with interest.
• Sentences that exceed two lines of text are taxes.
• Foreign-language phrases are taxes.
• Inside jokes are taxes.
• “As I’ll explain in a moment” is a tax with late fees.

Each tax is small. Individually, none is a crime. But they accumulate.

The Reader’s Budget

Every reader arrives at your piece with a finite budget — call it 1,500 cognitive calories — to spend on you. By the third long sentence, they’re at 800. By the fourth nested clause, they’re at 200. By the time you reach your hard-won conclusion, they’re broke.

They don’t leave because they disagree with you. They leave because they’re tired. The disagreement option requires energy they no longer have. Closing the tab is free.

This is the cruel asymmetry of public communication. You spent six hours writing it. They’re giving you six minutes. The economy has to balance, or the transaction collapses.

There’s a neurological reason this happens — and it’s connected to the way constant cognitive load degrades our highest-order thinking. The same Default Mode Network that gets shut down by overwork also gets shut down by dense, effortful reading. You’re not just losing a reader’s attention; you’re actively depleting the mental resource they’d need to engage with your idea.

Reducing the Tax

The fix is not to dumb things down. Rather, the fix is to identify which taxes are doing actual work and which are just habits. A useful audit:

• Cut every parenthetical. If the content matters, fold it into the main sentence. If it doesn’t, delete it.
• Replace “however” with “but.” Same logical connector, half the syllables.
• Define every acronym on first use, or replace it with the spelled-out form.
• Break long sentences in half. Two clauses? Two sentences.
• Cut every “that said,” “to be clear,” “of course,” and “needless to say.” They’re throat-clearing, not communication.
• Replace abstract examples with concrete ones. “A senior executive” is abstract. “A fifty-something CFO who can’t quit Excel” is concrete. Concrete is cheaper to process.

These are small surgeries, but the cumulative effect is dramatic. A piece that scored a 9-out-of-10 on the cognitive load scale becomes a 4. The same ideas — half the cost.

4. Why “Basic” Advice Outperforms Brilliance

There’s a creator on Twitter — you know the type — who posts daily observations so obvious they make you embarrassed for the species. “Your dreams won’t chase themselves.” “Sleep is underrated.” “Discipline is freedom.”

The intelligent person reads this and feels a primal, full-body rage. But that basic creator is doing something mathematically sophisticated, and it’s worth examining honestly.

Fluency as Truth

Cognitive psychologists have documented, repeatedly, that fluently-processed information is perceived as more true. The same fluency heuristic that makes us trust fast-talking experts makes us trust simple sentences. Easy-to-process content gets the benefit of the doubt; hard-to-process content gets the burden of proof.

When a creator distills an idea to ten words, they are operating at the maximum fluency the language allows. The reader processes the sentence in 200 milliseconds. The brain, having found the input cheap, codes it as truthful.

This is not a defect in the audience. It’s a feature of cognition. Fight it as a writer at your peril.

The Compression Skill

Reducing an idea to ten words is harder than expanding it to a thousand. Anyone can be verbose. Compression is the rare and valuable skill — the same skill that distinguishes a good poet from a competent one, a great copywriter from a corporate one.

The “basic” creator who repeatedly hits ten-word truths is not a shallow thinker who got lucky. They are a person who has practiced compression to a level most academics could not match if they tried.

The lesson is not to imitate the surface — posting trite aphorisms. The lesson is to practice the skill: take your sophisticated idea and ruthlessly compress it. Find the version that fits in one sentence. Find the version that fits in five words. The compressions will sometimes feel reductive, but they will travel.

Clarity Is Intelligence Performed in Public

Here is the reframe most intelligent writers need: Clarity is not the opposite of intelligence. Clarity is intelligence performed in public.

The genius who writes 4,000-word essays of dense brilliance has done the thinking. They have not done the translation. And translation is at least half the job — arguably more.

When you read a piece that takes a hard idea and lands it cleanly in your head on the first read, you are watching someone perform a feat of intellectual labour that most writers cannot. They have done the thinking and the translation. The translation is invisible, which is precisely why it looks effortless. It is, in fact, the harder of the two stages.

If you want a structured system for developing exactly this skill — the ability to make complexity legible — the AI-Resistance Self-Assessment is the diagnostic I’d run first. It was built from a year of research with 312 professionals whose roles were replaced by AI. The patterns that separated the 28% who recovered from the 72% who didn’t map directly to the communication and clarity skills this article covers.

5. The One-Variable Rule

The most useful operating principle for spreading complex ideas is this: You can only change one variable at a time.

How the Rule Works

Every piece of communication has multiple “variables” — things the reader has to absorb. Common variables include:

• The topic (what is this about?)
• The vocabulary (what specialised terms appear?)
• The format (how is this presented? Essay? List? Diagram?)
• The frame (what worldview does this assume?)
• The conclusion (what is the writer claiming?)

Change one of these from the reader’s expectations, and the piece is digestible. Change two, and it doubles in difficulty. Change three, and you’ve lost most of your audience. Change four, and you’re writing for nobody.

Why This Kills Brilliant Pieces

Most failed communication is just this — too many novel things at once. Consider the brilliant founder who tries to explain a novel business model (new topic) using novel jargon (new vocabulary) in a novel format (new presentation). Or the academic who introduces a new theory (new conclusion) within a new framework (new frame) using a new vocabulary.

These pieces are not failing because the ideas are too sophisticated. They’re failing because the writer tried to teach four lessons in the time the reader budgeted for one.

The same principle applies in professional communication. Information architecture for non-technical audiences comes down to exactly this: how many new things are you asking someone to hold in their head at once? The answer should usually be one.

Applying the Rule

When you have a complex idea you want to spread, ask:

• What is the new thing here? Pick one. Just one.
• What can I make familiar? Wrap the new thing in everything else the reader already knows — a familiar topic, familiar vocabulary, a familiar format.
• Where am I changing variables I didn’t have to? Often, writers introduce novelty in places that don’t need it. Strip out unnecessary novelty.

The piece should change one thing about the reader’s mental model. Everything else should be wallpaper — comfortable, familiar, free of cognitive load.

6. The 80% Rule

Here’s the part that stings for intelligent people to hear: You don’t get to teach the whole truth. Not in one piece, not on the internet, not in any form a normal human will read voluntarily.

Instead, you get to teach 80% of the truth. The remaining 20% — the edge cases, the exceptions, the “well, technically” — has to be sacrificed, deferred, or hidden in a footnote that 5% of readers will follow.

Why This Feels Like Lying (But Isn’t)

To the careful thinker, simplifying feels like betrayal. The exceptions are real. The edge cases matter. The qualifications are not optional — they’re load-bearing parts of the actual claim.

Except: the reader who walks away with 80% of your idea, intact and memorable, can find the other 20% on their own when they need it. Meanwhile, the reader who walks away with 0% because you tried to give them everything has been served by no one.

The whole truth, badly delivered, is a kind of vanity. It says I would rather be technically correct than actually useful. That is not, ironically, a very intelligent trade.

Choosing the Right 80%

Not all 80%s are equal. The skill lies in choosing the load-bearing 80% — the part of the idea that can survive without the qualifications.

A useful test: imagine your reader trying to apply your idea in a real situation. Which parts of your piece do they need to make the right call 80% of the time? That’s your core. The other 20% covers the cases where they will, with enough engagement, eventually go looking for more.

Strip your piece down to the 80% that does the most work. Trust your reader to find the rest when they need it.

7. The Dinner Party Test

Before you publish anything — an essay, a memo, a strategy doc, a tweet — run it through this test.

Imagine someone reads your piece. They like it. A week later, at a dinner party, they try to explain your idea to a stranger. They get half the details wrong. They mangle one of your favourite sentences. They forget the third bullet point entirely.

Does the core idea still survive that mauling?

If yes, you wrote a real idea. It was small enough, sharp enough, memorable enough to pass from one head to another without losing its shape. It is, in the truest sense, contagious.

If no — if your idea only works when delivered exactly the way you delivered it, with all caveats present and accounted for — then you did not write an idea. You wrote a lecture. And lectures don’t travel.

The Test in Practice

The Dinner Party Test is not a metaphor. Try it literally. Tell a friend or family member about something you’ve written, then a week later ask them to explain the idea back to you. The version that comes back is the version that actually exists in the world.

Most writers, doing this for the first time, are horrified. The piece they spent six hours crafting comes back as a rough sketch, missing two-thirds of the qualifications and several of the supporting examples. That can’t be right, they think. They must have just forgotten the rest.

But the rest didn’t just get forgotten. It never made it across in the first place. You were tapping. They were hearing thuds. The version that survives is the version that matters.

8. A Framework for Translating Hard Ideas

Putting it all together, here is a complete framework for taking a sophisticated idea and translating it into a form that spreads.

Step 1: Find the Single Sentence

Before you write anything, force yourself to express the entire idea in one sentence. Not a paragraph. Not a thesis statement. One sentence. If you can’t, you don’t yet have an idea. You have a collection of related thoughts. Keep working until you can compress.

Step 2: Pick the Anchor Story

Every spreadable idea needs a story attached to it. The story is not decoration — it’s the delivery mechanism. Your reader will remember the story long after they’ve forgotten the framework.

Pick a single concrete anecdote that makes the idea feel inevitable. The story should be specific (real names, real places, real numbers when possible) and should require the idea to make sense of it.

Step 3: Identify the One Variable You’re Changing

What is the new thing you’re asking the reader to absorb? Name it explicitly, in your notes. Then make sure everything else in the piece is familiar — familiar topic territory, familiar vocabulary, familiar format.

Step 4: Cut to the Load-Bearing 80%

Write the piece, then ruthlessly identify which paragraphs are doing the work and which are decoration or qualification. Cut the decorations. Move the qualifications to a footnote, an FAQ, or a follow-up piece.

Step 5: Apply the Clarity Tax Audit

Read the piece aloud. Every time you stumble or have to re-read a sentence, that sentence is too expensive. Rewrite it shorter. Replace abstract nouns with concrete ones. Break long sentences in half. Replace jargon with plain language unless the jargon is genuinely irreplaceable.

Step 6: Run the Dinner Party Test

Show the piece to one person who is not in your field. Ask them to explain the idea back to you a few hours later. Note what survives and what doesn’t. The parts that don’t survive usually need to be either cut or rewritten.

Step 7: Cut Another 20%

After the test, cut another 20% of the length. This is almost always possible. Most writers’ first drafts are longer than they need to be. Most published pieces are still longer than they need to be. The version closest to the spreading-shape is shorter than feels comfortable.

9. Common Objections (and Why They’re Traps)

Intelligent writers have predictable defences against this entire framework. Each defence sounds respectable, and each is, on closer inspection, a way of preserving the comfort of writing for the imagined panel of peers instead of the actual audience.

“But my topic is genuinely complex.”

Everything is genuinely complex. Quantum mechanics is complex; nevertheless, Brian Greene wrote The Elegant Universe and sold a million copies. Evolutionary biology is complex; yet Richard Dawkins wrote The Selfish Gene and reshaped public understanding. The complexity of your topic is not a licence to write impenetrably. It’s an invitation to work harder on the translation.

“My audience is sophisticated. They want depth.”

Sophisticated audiences can spot effort. They can spot when you’ve earned a complex sentence by laying the groundwork for it. What they can’t tolerate — and won’t forgive — is unnecessary opacity. The smartest readers are the most ruthless about wasted words. They are not impressed by your jargon. They are bored by it.

“I don’t want to oversimplify and mislead people.”

This is the most respectable-sounding excuse and the most lethal. Almost every “oversimplification” you’re afraid of is, in fact, just a simplification. It’s not wrong; it’s incomplete. And incomplete-but-clear beats complete-but-incomprehensible every single time, because incomplete-but-clear is the only version anyone will read.

“Plain writing won’t earn me credibility in my field.”

Two responses. First: in any field worth being in, the most respected practitioners are usually the ones who can explain complex ideas to outsiders — Feynman in physics, Krugman in economics, Sean Carroll in cosmology. These are not lesser thinkers; they are thinkers who paid the additional cost of translation. Second: if your field genuinely punishes clarity, that is important information about your field, and you may want to consider whether it’s the right pond.

“Short pieces feel shallow.”

Short pieces can be shallow. They can also be the most concentrated form of a deep idea. The difference is not length; it’s the density of useful thought per word. A 4,000-word essay can be shallow if it pads the same idea twenty different ways. A 600-word piece can be deep if every sentence carries weight.

10. Key Takeaways

• The Curse of Knowledge prevents experts from imagining what their audience does not know — the song in your head is silent on the table.
• Intelligent writers compound the problem by preferring nuance over resonance and writing to perform expertise rather than to communicate.
• The Clarity Tax is the cumulative cognitive cost you charge your reader; even small charges add up to bankruptcy.
• “Basic” advice spreads not because audiences are shallow but because fluency is perceived as truth, and compression is a sophisticated skill.
• The One-Variable Rule: change only one element of the reader’s mental model per piece. Wrap novelty in familiarity.
• The 80% Rule: prioritise the load-bearing core that can survive without qualifications. Trust the reader to find the rest.
• The Dinner Party Test: an idea only counts as transmitted if it survives being explained badly by someone else.
• Translation is not a lesser form of intellectual work. It is, often, the harder form.

This video covers the core of the article in 60 seconds:

Extended 8-minute breakdown: youtu.be/zaHAAx5FiMg

11. Frequently Asked Questions

What is the Curse of Knowledge?

The Curse of Knowledge is a cognitive bias, first formally described in a 1989 economics paper and popularized by Made to Stick, in which a person who knows something cannot accurately imagine what it’s like not to know it. As a result, they unconsciously assume their audience shares their context, leading to systematic communication failures even when both parties are intelligent and engaged.

Why do my long, well-researched posts get less engagement than simple ones?

Two reasons. First, the fluency heuristic — the brain treats easy-to-process content as more credible. Second, the cognitive budget problem: readers have a finite amount of attention to spend on any single piece, and complex pieces exhaust that budget before reaching the conclusion. The fix is not to dumb your ideas down but to compress them and reduce unnecessary friction.

How do I know if I’m writing too densely?

Read your piece aloud. Note every place you stumble, lose your breath, or have to re-read a sentence to parse it. Each of those is a place your reader will also stumble — except they won’t have your context, so it’ll be worse for them. Rewrite those sentences to be shorter, clearer, and more concrete.

Doesn’t simplification mean dumbing down?

No. Simplification means removing what’s not essential; dumbing down means removing what is essential. The former requires more skill than dense writing; the latter requires less. The two outcomes look similar from the outside but feel completely different to read.

What if my audience really does want technical depth?

Some audiences do. Specialised academic journals, technical documentation, internal engineering memos — these have their own conventions, and their readers come prepared for density. However, almost all public writing addresses an audience that hasn’t done the prerequisite work. If you’re writing on a blog, on social media, or for a general business audience, you are almost never writing for the technically prepared reader, even when it feels like you are.

How do I balance accuracy and accessibility?

Use the 80% rule. Identify the load-bearing core of your idea — the version a reader needs to make the right call most of the time. Cover that thoroughly and clearly. Then move the qualifications, edge cases, and exceptions into footnotes, FAQs, or follow-up pieces. Trust your reader to seek out the additional 20% if their use case requires it.

How long should a piece be?

There is no single right answer, but most pieces are too long. A useful test: if you cut your piece by 30%, does it become worse, or does it become more readable while saying the same thing? Most writers find the latter happens far more often than they’d expect. A piece’s natural length is almost always shorter than the writer’s first instinct.

Can I learn to write more clearly, or is it a fixed skill?

Clarity is one of the most trainable skills in writing. The two most useful practices: (1) reading your work aloud and rewriting every place you stumble, and (2) compressing — taking a piece you’ve written and cutting it to half its length while preserving the core argument. Both practices, done weekly for six months, will transform most writers’ work.

If you want a structured implementation system — worksheets, frameworks, and a 30-day roadmap — the AI-Resistance Self-Assessment is the audit I’d pair with this. It maps your current career exposure against 312 real displacement cases, and gives you a concrete readiness score.

Take the AI-Resistance Self-Assessment → (£5 · 12 pages · 30 questions)

12. Further Reading

• Chip Heath and Dan Heath, Made to Stick: Why Some Ideas Survive and Others Die — the canonical book on what makes ideas spread.
• Steven Pinker, The Sense of Style — a cognitive scientist’s guide to clear writing in the 21st century.
• Daniel Kahneman, Thinking, Fast and Slow — for the underlying cognitive science of how readers process information.

If you found this useful, subscribe below for weekly essays on writing, persuasion, and the psychology of clear thinking. Share this piece with someone who needs to hear it.