the tapping study: why your explanation sounds clearer to you than to anyone else

The study

Elizabeth Newton submitted her PhD dissertation, The rocky road from actions to intentions, to Stanford's Department of Psychology in June 1990. Her question was simple: when we act, do we overestimate how easily other people can read what we mean?

To test it, she recruited 80 Stanford undergraduates and split them into pairs. In each pair, one person was the tapper and the other the listener. Tappers chose three tunes from a list of 25 well-known songs and tapped the rhythm with a finger while sitting back to back with their partner. Listeners wrote down their guess after each tune. Between songs, each tapper estimated how likely it was that the listener had got it right.

A second group of 80 students took part as informed observers. They were told the name of each tune, then heard the experimenter tap it, and were asked what share of an audience of 100 people would recognise it.

What it found

The tappers were confident. Their estimates averaged 50 percent, ranging from 10 to 95 percent. In reality, listeners named 3 tunes out of 120 attempts, a success rate of 2.5 percent. That figure sat below the lowest estimate any tapper gave.

The informed observers, who had never tapped or listened themselves, were just as confident. Knowing the name of the song was enough to make the tapping sound like music to them.

Newton's explanation is the part that has travelled. When you tap a tune you know, you hear the melody in your head. Your listener hears only knocks on a table. A held note in your mind is, to them, just a gap. You cannot easily switch off the tune and hear what they hear.

Most people now meet the study through Chip Heath and Dan Heath, authors of the book Made to Stick, who also wrote about the curse of knowledge in the Harvard Business Review in 2006, applying it to leaders whose strategy statements mean little to frontline staff. In a 2007 NPR conversation about the book, Chip Heath described the result as three songs guessed out of 120, which matches the results chapter of the dissertation.

Did it hold up?

There are three honest things to say.

First, the original was never published in a journal. As far as we can find, the tapping study exists as a dissertation chapter, not a peer-reviewed article, and it has not been through the scrutiny a journal paper receives. The numbers people quote come from the dissertation or from retellings of it. Even the source has a small inconsistency: the dissertation's abstract says listeners identified two of 150 tunes, while the results chapter and its table say 3 of 120. The NPR retelling dates the study to 1991; the dissertation is dated June 1990. None of this changes the story, but it is a reason to quote the numbers loosely.

Second, we could not find a published direct replication of the tapping task itself. The sample was 40 pairs of students at one university.

Third, the idea behind it is well supported, with limits. The phrase "curse of knowledge" comes from a 1989 paper by the economists Colin Camerer, George Loewenstein and Martin Weber in the Journal of Political Economy. They found that better-informed people could not set aside what they knew when predicting the judgements of less-informed people, even when it cost them money. Trading in a market reduced the bias by about half, but did not remove it.

Closer to everyday speech, Boaz Keysar and Anne Henly at the University of Chicago published a study in Psychological Science in 2002. Forty pairs took part. Speakers read ambiguous sentences aloud, meaning one particular sense, and then judged whether their listener had understood. When speakers believed they had been understood, they were wrong 46 percent of the time. Observers who knew the intended meaning, but had not spoken, did not show the same overconfidence. This is the tapping study's lesson in words rather than rhythms.

The effect is real but not always large. In 2007, Susan Birch and Paul Bloom reported that adults' own knowledge biased their judgements of what another person would believe. When Rachel Ryskin and Sarah Brown-Schmidt tested this with seven experiments and 3,074 participants in 2014, the effect was in the same direction but less than half the original size, and they suggested it may matter less in real life than first thought.

How it gets misused

• Treating 2.5 percent as a law. It was one task, with one group of students, in which the medium (a finger on a table) strips out nearly everything. Most of our explanations carry far more than a rhythm.
• "Experts cannot teach." The study says nothing of the kind. It says experts need to check, because their sense of clarity is not a reliable guide.
• Blaming the listener. The listeners were not slow. They had a fraction of the information the tapper had.

What it means for you

The tapping study is about anyone who explains their own work to someone outside it. A software lead describing a release to the sales team. A nurse explaining a discharge plan to a family in Doha. A founder in Kochi pitching to an investor who has never run a kitchen. A parent explaining a decision to a fourteen-year-old.

In each case, you hear the melody and they hear the taps. Three habits help.

Start with what it means for them, then the detail. The tune you hear is full of context. Give the listener the headline first, so they know what song they are listening to.

"The short version: the launch moves by a week, and your clients will not notice. Here is why."

Replace your inside words. Every team has terms that feel ordinary after a month and mean nothing to outsiders. Listen for them in your own sentences.

Ask them to play it back, without testing them. The Keysar and Henly speakers were wrong nearly half the time they felt understood. The only way to know is to ask, in a way that puts the burden on you.

"I have been inside this for weeks, so I may have skipped steps. What did you take away from that?"

"Tell me which part was least clear. That is my fault, not yours."

Sources

1. Elizabeth L. Newton, The rocky road from actions to intentions, PhD dissertation, Department of Psychology, Stanford University, June 1990 (UMI order no. 9102S21; scanned copy hosted by gwern.net). www.gwern.net/doc/psychology/cognitive-bias/illusion-of-depth/1990-newton.pdf
2. Chip Heath and Dan Heath, The Curse of Knowledge, Harvard Business Review, December 2006. hbr.org/2006/12/the-curse-of-knowledge
3. NPR Morning Edition, The Secret Behind Why Ideas Stick, with Chip and Dan Heath, 19 February 2007 (via WFAE). www.wfae.org/2007-02-19/the-secret-behind-why-ideas-stick
4. Colin Camerer, George Loewenstein and Martin Weber, The Curse of Knowledge in Economic Settings: An Experimental Analysis, Journal of Political Economy 97(5), 1232-1254, 1989 (author copy at Carnegie Mellon). www.cmu.edu/dietrich/sds/docs/loewenstein/CurseknowledgeEconSet.pdf
5. Boaz Keysar and Anne S. Henly, Speakers' Overestimation of Their Effectiveness, Psychological Science 13(3), 207-212, 2002. doi.org/10.1111/1467-9280.00439
6. University of Chicago Chronicle, report on Keysar and Henly's study of speakers' overconfidence, 9 May 2002. chronicle.uchicago.edu/020509/keysarresearch.shtml
7. Susan A. J. Birch and Paul Bloom, The Curse of Knowledge in Reasoning About False Beliefs, Psychological Science 18(5), 382-386, 2007. doi.org/10.1111/j.1467-9280.2007.01909.x
8. Rachel A. Ryskin and Sarah Brown-Schmidt, Do Adults Show a Curse of Knowledge in False-Belief Reasoning? A Robust Estimate of the True Effect Size, PLOS ONE, 2014. journals.plos.org/plosone/article?id=10.1371/journal.pone.0092406

Published by The Full Stop, the journal of Brightside English, in the The Experiments series. Every factual claim links to a source below; where accounts or findings are disputed, we say so. Sources checked on 26 Oct 2026. Reviewed by a Brightside senior coach. Corrections: see our editorial standard.