The Curse of Knowledge

Once you know something, you can no longer imagine not knowing it. And this inability makes you systematically overestimate what others understand — in your explanations, your writing, your presentations.

Origin

The concept was formalized by several researchers. Colin Camerer, George Loewenstein, and Martin Weber (1989) documented it in The Curse of Knowledge in Economic Settings. But it was Elizabeth Newton’s experiment in her 1990 Stanford doctoral thesis that made the phenomenon concrete and memorable.

Chip Heath and Dan Heath popularized the concept in Made to Stick: Why Some Ideas Survive and Others Die (2007), establishing it in the vocabulary of communication and marketing.

The Theory

Newton’s experiment is simple and striking. Participants — the “tappers” — tap the rhythm of a well-known song (like “Happy Birthday”) on a table. Other participants — the “listeners” — must guess the song.

Before the experiment, tappers estimate that 50% of listeners will guess correctly. The reality: 2.5% — one in forty.

Why the gap? The tapper hears the melody in their head while tapping. For them, the rhythm is obvious — it is inseparable from the song. The listener hears only: tap-tap-tap. Without the melody, the rhythm means nothing.

The tapper cannot “un-hear” the melody. They can no longer put themselves in the position of someone who doesn’t hear it. That is the curse of knowledge: once acquired, knowledge becomes invisible to the one who holds it.

In Practice

In teaching: Experts forget what it was like not to know. They skip “obvious” steps that are only obvious to them. They use jargon without defining it. They underestimate the time needed to learn.

In communication: A technical writer who knows their product intimately assumes context their reader doesn’t have. A presentation built by experts for beginners loses its audience by slide two.

In project management: The person who “has the vision” forgets that their team only sees tasks. What is obvious to them is not obvious to those without their level of information.

Practical application: Rereading your own writing in “would someone who knows nothing understand this?” mode is harder than it sounds. The bias is active at every reading — you cannot un-learn.

The most reliable solution remains external: have someone without the context read it. Non-expert feedback is often the best indicator of clarity.

Nuances and Limits

The curse of knowledge is not a personal flaw — it is a structural cognitive bias. You cannot simply “try to pretend you don’t know,” because the bias operates below the conscious level.

It does not imply that experts necessarily communicate poorly — some develop strategies to work around it. But these strategies are learned and actively maintained; they are not natural.

The phenomenon relates to hindsight bias (in retrospect, we overestimate what we knew beforehand), the illusion of transparency (believing our thoughts are more visible to others than they are), and inversely to the Dunning-Kruger effect (beginners overestimate their competence, experts underestimate theirs).

The Feynman Technique is one method for actively countering this bias: by forcing simple explanation, you detect areas where knowledge has become too implicit.

Sources: Newton, E. (1990). PhD dissertation, Stanford University · Camerer, Loewenstein & Weber (1989). Journal of Political Economy · Heath, C. & Heath, D. (2007). Made to Stick