What is the cocktail party effect?
The phrase covers a 1953 problem about following one voice and a 1959 finding about hearing your own name. The studies behind each are not the same.
Answer to the question
What is the cocktail party effect?
The cocktail party effect is the ability to follow one voice among several, together with the finding that some listeners notice their own name in a voice they are ignoring. Colin Cherry posed the problem in 1953, and Neville Moray reported the name result in 1959. In the studies since, roughly one listener in three notices.
Key takeaways
- Cherry (1953) posed the cocktail party problem; in his experiments a change from a male voice to a female voice was nearly always noticed in the ignored message, and a switch from English to German usually was not.
- Moray (1959) found that only a listener's own name reliably broke through from an ignored message; about a third of his listeners reported hearing it.
- Conway and colleagues (2001), who give a figure of 34.6 per cent for Wood and Cowan's 1995 revisit, found 65 per cent of 20 low-span students against 20 per cent of 20 high-span students.
- Broadbent (1958) proposed an early filter on physical features; Treisman (1960) proposed that the ignored message is turned down rather than blocked.
- Shotton uses the cocktail party effect to argue for relevance in advertising; that is an inference from the laboratory work, not one of its results.
The cocktail party effect names two different things, and the first is not an effect at all. It began in the early 1950s as an engineering question about pulling one voice out of several.
Colin Cherry put that question in a 1953 paper and called it the cocktail party problem. The second meaning arrived six years later, when Neville Moray reported that a listener ignoring one voice can still catch their own name in it. Popular accounts fold the two together under one label. They answer different questions and rest on different experiments.
What Cherry Asked and What Listeners Kept
Cherry's 1953 paper in the Journal of the Acoustical Society of America used the phrase for the difficulty of following one speaker while others talk at once. The paper's abstract describes objective recognition experiments with continuous speech, analysed mainly by statistics. There are two kinds of test: one in which a listener hears two speech signals at once, and one in which a different message goes to each ear. That much of the design is on the record.
Textbook accounts of the paper add the results. Two messages from the same talker, mixed and played to both ears, were hard to separate. With a different message in each ear, listeners could choose which to follow.
The cocktail party effect, in its first sense, is therefore the ability to follow one voice in a noisy room. Researchers file it under selective attention, the capacity to favour some inputs and leave others unprocessed.
The experiments that gave the paper its reputation used what became known as shadowing: each ear heard a different message and the listener repeated one of them aloud as it played. The standard account is that listeners followed the chosen message accurately and retained almost none of the content of the other.
On the same account, some physical features of the rejected voice got through and meaning did not. The changes made partway through the ignored message are usually reported like this, and the detail is the part of the story to hold most loosely.
| Change in the ignored message | Reported as noticed? |
|---|---|
| Male voice to female voice | Yes |
| Speech to a pure tone | Yes |
| English to German | Usually not |
Sound survived the journey and meaning did not. That pattern is the part of the cocktail party effect that Cherry's work is credited with.
Why a Filter Could Not Admit Names
Donald Broadbent turned that pattern into a theory in his 1958 book Perception and Communication. In filter theory, a message is selected early on its physical features, such as the ear it arrives in, and the rejected message is blocked before its meaning is analysed. Cherry's results fit this account well, and for a few years the cocktail party effect was understood as a story about an early filter.
Moray's 1959 experiment asked what, if anything, crosses the filter. Listeners followed one message and ignored another. A short list of words repeated many times in the ignored message left no trace on a later recognition test, and telling listeners to remember numbers hidden in it did not help them recall more of those numbers. Their own name was the exception. Moray concluded that only material of personal importance to the listener breaks through.
A filter that blocks meaning cannot let a name in. Anne Treisman proposed a revision in 1960, now called attenuation theory: the rejected message is turned down rather than shut out, and words that matter to the listener keep a low threshold, so they can reach awareness at low volume.
The two accounts differ on one point, whether the meaning of an ignored message is analysed at all. Filter theory says it is not. Attenuation theory says it is, at reduced strength, which is what lets an important word through. Both remain in the textbooks, and the evidence between them is mixed.
How Many Listeners Notice Their Own Name?
The name result is often told as if everyone hears their name across a room. The measured figures are lower, and the cocktail party effect in this second sense is a minority result. About a third of the listeners in Moray's study reported hearing their name in the ignored message. Noelle Wood and Nelson Cowan revisited the question in 1995, and a later paper by Cowan and colleagues, described below, gives their figure as 34.6 per cent.
A 2001 study by Andrew Conway, Nelson Cowan and a colleague tested who the noticers were. Forty undergraduates were split into two groups of 20 by their score on a memory span task, which measures how much a person can hold in mind while doing something else. In the task, participants repeated a message in one ear while ignoring a different message in the other, with their own first name inserted into the ignored message. Afterwards they were asked whether they had heard anything unusual.
Span here was scored on an operation span task, in which participants solve arithmetic problems while remembering unrelated words. Of the low-span group, 65 per cent reported hearing their name in the ignored message, against 20 per cent of the high-span group. The authors read the pattern as a sign that people with lower capacity are worse at blocking distraction.
The tasks matter when reading these numbers. Shadowing asks a listener to commit to one message through headphones, which a real party does not. On our reading, the rates describe how often a name breaks through under that commitment, and they should not be carried over to a room full of people without further evidence.
A preregistered replication by Jan Philipp Röer and Cowan in 2021 confirmed that one's name captures attention and found that unexpected words do not. The cocktail party effect holds up as an average and describes a minority of listeners. The follow-up on capacity, and on what happens in older listeners, is covered in our article Why You Hear Your Name in a Crowd.
From the Laboratory to Advertising
Richard Shotton gives the cocktail party effect a chapter in The Choice Factory. He argues that personal relevance is what gets a message past the brain's filter, so relevance is the advertiser's best route to being noticed. The argument runs in two steps: relevant material can reach awareness, and advertising benefits when it is relevant. Only the first step belongs to the studies, and it concerns speech heard in a laboratory. The second is a marketing inference from the cocktail party effect, not a result of it.
Standing out is a separate route to being noticed, one that works on the stimulus itself instead of on the listener's concerns, and we cover it in Why the Odd One Out Gets Remembered. Shotton's own practical advice is to localise appeals rather than personalise them with a viewer's name, which swaps a cue the studies tested directly, the listener's own name, for a broader kind of relevance, and we examine that choice from the name side, including what the laboratory results do and do not license, in Why Your Name on a Product Works.
Common questions
What is the cocktail party effect?
The cocktail party effect is the ability to follow one voice among several, together with the finding that some listeners notice their own name in a voice they are ignoring. Colin Cherry posed the problem in 1953, and Neville Moray reported the name result in 1959. In the studies since, roughly one listener in three notices.
Who discovered the cocktail party effect?
Colin Cherry described the cocktail party problem in 1953 and studied how listeners separate two messages. The finding that a listener can notice their own name in an ignored message is Neville Moray's, from 1959.
Does everyone hear their own name in a crowd?
No. About a third of listeners did in the classic studies, and in a 2001 study those with lower memory span noticed more often than those with higher span.
Sources
Shotton, R. (2018). The Choice Factory: 25 behavioural biases that influence what we buy. Harriman House. Seed book; vault hub Books/The Choice Factory.
Cherry, E. C. (1953). “Some experiments on the recognition of speech, with one and with two ears.” Journal of the Acoustical Society of America, 25, 975-979. doi:10.1121/1.1907229.
Broadbent, D. E. (1958). Perception and Communication. Pergamon Press. doi:10.1037/10037-000
Moray, N. (1959). “Attention in dichotic listening: affective cues and the influence of instructions.” Quarterly Journal of Experimental Psychology, 11(1), 56-60. doi:10.1080/17470215908416289.
Treisman, A. M. (1960). “Contextual cues in selective listening.” Quarterly Journal of Experimental Psychology, 12(4), 242-248. doi:10.1080/17470216008416732
Wood, N., & Cowan, N. (1995). “The cocktail party phenomenon revisited: how frequent are attention shifts to one’s name in an irrelevant auditory channel?” Journal of Experimental Psychology: Learning, Memory, and Cognition, 21(1), 255-260. doi:10.1037/0278-7393.21.1.255.
Conway, A. R. A., Cowan, N., & Bunting, M. F. (2001). “The cocktail party phenomenon revisited: the importance of working memory capacity.” Psychonomic Bulletin & Review, 8(2), 331-335. doi:10.3758/BF03196169.
Röer, J. P., & Cowan, N. (2021). “A preregistered replication and extension of the cocktail party phenomenon: one’s name captures attention, unexpected words do not.” Journal of Experimental Psychology: Learning, Memory, and Cognition, 47(2), 234-242. doi:10.1037/xlm0000874.