Music and work performance: what do we know?

Many people listen to some kind of music while working, or perhaps to some kind of colored noise. Many people will swear that this works. There are multiple ways it can work:

  1. Increases work performance.
  2. Increase work enjoyment (positive mood effect).
  3. Makes work seem to go faster (time experience effect).

They are not necessarily causally independent paths. Perhaps (2) causes (3). Perhaps (2) causes lower (3) due to distraction.

Anecdotally, we would probably say that (1) is likely to be true, especially when the music is synchronized to bodily movements. Otherwise, why would the military bother with march music? Still, just because it works with physical labor, does not mean it works with mental labor too.

What science has been done so far?

Given that hundreds of millions of people probably use music while working, it would seem like an obvious thing to research. But when I looked, I only found a few studies. So, this topic goes on the list of things that are obvious to research and has potentially large practical applications, but which scientists for some reason have not researched. Instead they wasted lots of time and money studying irreproducible fads.

But I was able to find some studies:

This is the relevant studies on the first two pages on Google Scholar. So, nothing very convincing. It looks like yet another fragmented, probably publication bias-ridden literature of little practical use. Science will have to mostly start over and begin with larger, pre-registered studies with public data. Preferably, these should also be replicated across research teams.

What would I do?

Suppose that someone gave me some students/people to command and some money, what would I do?

Something like this:

  • Pre-register analyses so others can believe that didn’t cheat. Have someone skeptical and competent (e.g. Zigerell) look them over before collecting data. Share all materials.
  • Recruit about N=100 persons.
  • Within subject design (for higher statistical power relative to the sample size).
  • Two outcome measures: easy and hard tasks. E.g. something simple like reading a simple/complex text and answering comprehension questions. Operationalization: count the number of correct answers.
  • A couple of music conditions:
    • silence (using active noise-cancelling headphones perhaps)
    • white noise
    • instrumental electronic music, e.g. one of Deadmau5’s instrumental songs
    • vocal electronic music, e.g. one of Deadmau5’s vocal songs
    • metal music, e.g. early Metallica
    • pop music, e.g. Justin Bieber/Miley Cyrus
  • There should be the same number of unique texts (one in each difficulty) as music conditions. And they should be randomly combined in a random person at the individual-level.

Then we simply compare across the conditions.


Gwern informs me of his prior post on the topic. He managed to find a meta-analysis, which abstract reads:

Background music has been found to have beneficial, detrimental, or no effect on a variety of behavioral and psychological outcome measures. This article reports a meta-analysis that attempts to summarize the impact of background music. A global analysis shows a null effect, but a detailed examination of the studies that allow the calculation of effects sizes reveals that this null effect is most probably due to averaging out specific effects. In our analysis, the probability of detecting such specific effects was not very high as a result of the scarcity of studies that allowed the calculation of respective effect sizes. Nonetheless, we could identify several such cases: a comparison of studies that examined background music compared to no music indicates that background music disturbs the reading process, has some small detrimental effects on memory, but has a positive impact on emotional reactions and improves achievements in sports. A comparison of different types of background music reveals that the tempo of the music influences the tempo of activities that are performed while being exposed to background music. It is suggested that effort should be made to develop more specific theories about the impact of background music and to increase the methodological quality of relevant studies.

So, this reinforces my conclusion of a fragmented literature. They speculate about moderators effects, but moderator effects analyses are unreliable (I’m not familiar with any hard evidence but considering what LASSO does to normal OLS, it would eviscerate typical meta-regression findings), so I wouldn’t put much weight in that.