Baader-Meinhof Phenomenon: Why a New Word Seems to Be Everywhere

The experience has two parts. Zwicky's 2005 post named the first, selective attention; later accounts added the second, confirmation bias. Neither requires anything unusual in the world.

Selective attention does the first part. Far more language passes a person in a day than can be processed, and the mind filters on relevance, novelty and personal meaning. A word with no entry in memory has no relevance to attach to, so it is read over or heard past without leaving a trace of having been encountered. Learning the word creates the entry, and from then on the same word, at the same frequency, catches. The demo gives a taste of this half: a word appeared three times before it was named, and a count other than three shows only that the reported count did not match. The word may have gone unattended, or been read and not counted, or been counted and misremembered by the time the page asked; the demo does not separate these. Inattentional blindness experiments make the general point more dramatically; in Simons and Chabris's 1999 study, about half of the people counting basketball passes failed to notice a person in a gorilla suit walk through the scene.

Confirmation bias does the second part. Once the word is being noticed, each sighting is logged as evidence for the impression that it has become common, while the much larger number of occasions on which the word does not appear are not logged at all, because an absence is not an event. The tally in memory is one-sided by construction, so the impression of a surge grows with each hit and is never corrected by a miss. Add the ordinary human readiness to find patterns in coincidence, and two sightings in 24 hours feel like a message.

No experiment has measured the Baader-Meinhof phenomenon as such; it is a name for a reported experience, not a laboratory effect with an effect size. What is well established is each of its components: that unattended stimuli are frequently not perceived, that newly learned items are more easily detected, and that people count confirming instances more readily than disconfirming ones. The phenomenon is what those three facts feel like from the inside.

There is no original study. There is an original letter. The first use the Oxford English Dictionary records is in the Bulletin Board column of the St. Paul Pioneer Press, a reader-contribution column, on October 16, 1994, where a writer announced a phenomenon broad enough to cover many forms of eerie coincidence and added, "I have dubbed it The Baader-Meinhof Phenomenon." He signed as Gigetto on Lincoln and sent a series of such letters. The group itself, the West German militant organization of the 1970s, has nothing to do with the effect. The experience reached a wide audience through a 2006 article on the website Damn Interesting, and in a message quoted there that October the writer gave his name as Terry Mullen of St. Paul and said he had been trying to amuse his friends rather than coin a term.

The scholarly name came separately. On August 7, 2005, the Stanford linguist Arnold Zwicky published a post on Language Log, "Just between Dr. Language and I," about the belief that certain usages were new when they were in fact old. He named two illusions in it. The recency illusion is the belief that something you have noticed only recently is in fact recent. The frequency illusion is the companion: once you have noticed a phenomenon, you think it happens a great deal, even all the time. Zwicky called both selective attention effects: once you have noticed a phenomenon, he wrote, your estimate of its frequency is skewed by your noticing nearly every occurrence that passes you. He was describing a hazard for people who study language, but the description fit the Pioneer Press letter exactly, and the two names have been treated as synonyms since.

In March 2019 the Oxford English Dictionary added both, defining the frequency illusion as a quirk of perception whereby a phenomenon to which one is newly alert suddenly seems ubiquitous and cross-referencing the Baader-Meinhof phenomenon as its other name. A Language Log post by Ben Zimmer on March 19, 2019, marked the entry and the term's origin on that site.

The pattern appears whenever something becomes personally relevant and is then met again. The rows below are illustrations, not findings from the sources cited on this page.

The phenomenon is harmless most of the time, and pleasant; the counters matter when the sense of a surge becomes an argument about the world.

• Ask when you learned the term. If the sightings began the week you learned it, the simplest explanation is that your filter changed, not the world.
• Count the misses. For one day, notice the forecasts, pages or conversations in which the term did not appear. The one-sided tally in memory needs the other column filled in.
• Check the frequency independently. A dated search, a usage corpus, sales figures or a trend chart will say whether the thing has actually become more common. Sometimes it has, and the feeling is tracking something real.
• Separate noticing from meaning. Two sightings in a day are what frequent things do at random; a feeling of significance is not evidence of a pattern.
• Use it on purpose. The same mechanism that makes a new word jump out is what makes vocabulary stick when learning a language; attention primed by a label is attention that learns.

The phenomenon sits among several biases of attention and tallying, and it is often conflated with them.

• It is not a separate effect from the frequency illusion; the two are the same experience under two names, the popular one from a 1994 newspaper letter and the scholarly one from Zwicky's 2005 post. This page is about the name and the lived experience; the frequency illusion page treats the attention mechanism in more depth and the two link each other.
• It is not the recency illusion, Zwicky's other coinage, the belief that something you have only recently noticed is in fact new. The frequency illusion is about how common something seems; the recency illusion is about how old it is. They often travel together.
• It is not confirmation bias in general, though it draws on it. Confirmation bias is the broad tendency to seek and weigh evidence that fits a belief; the one-sided count of sightings is one instance.
• It is not attentional bias as the term is used in clinical research, where it names the way attention is drawn to threat-related or mood-related stimuli. Baader-Meinhof is about novelty and relevance, and it applies to neutral words and station wagons.