The Meme:

A counter view:

This idea seems largely meaningless to me. I don't know if Pavlovian conditioning to a stimulus counts as a habit, but if so, it can be done in minutes.

My question is where is this meme from? Is there, at its base, any research (as at least one of them claimed)?

  • I can get in a habit in a few days Feb 11 '12 at 16:58
  • @ratchet You can't call it a habit, if you've had it less than 21 (or n > a few) days :)
    – UncleBens
    Feb 11 '12 at 21:35
  • I call a habit something I do regularly and am uncomfortable not doing (for no real reason), I can get into that very easily (hell my habit of opening SE on a new window came in less than a week and remains until now for over 8 months now Feb 11 '12 at 22:43
  • some drugs can be addictive with one or a very few doses (think heroin). I'd call BS without a very strict definition of "habit".
    – jwenting
    Feb 12 '12 at 7:47
  • 1
    Quick ironic disproof: psychology-related data is always statistical and has big dispersion, and it's extremely unlikely to have it ~5%. So anyone who claims "21 days" (not "about 20 days" or "3 weeks") and doesn't provide dispersion has probably done his research in a wrong way.
    – Steed
    Mar 25 '13 at 11:43

No, the 21-day habit rule is pseudo-scientific.

First, it can take much longer:

These researchers asked 96 volunteers to form a daily habit, over a 12 week period, and record their progress.

The majority (82) of participants provided sufficient data for analysis, and increases in automaticity (calculated with a sub-set of SRHI items) were examined over the study period. Nonlinear regressions fitted an asymptotic curve to each individual's automaticity scores over the 84 days. The model fitted for 62 individuals, of whom 39 showed a good fit. Performing the behaviour more consistently was associated with better model fit. The time it took participants to reach 95% of their asymptote of automaticity ranged from 18 to 254 days; indicating considerable variation in how long it takes people to reach their limit of automaticity and highlighting that it can take a very long time. Missing one opportunity to perform the behaviour did not materially affect the habit formation process.

The source of the myth has been traced by Guardian columnist Oliver Burkeman who writes:

We probably owe this particular example of pop-psychology wisdom to Maxwell Maltz, the plastic surgeon who wrote the 60s bestseller Psycho-Cybernetics. He claimed to have observed that amputees took an average of only 21 days to adjust to the loss of a limb. Therefore, he reasoned – deploying the copper-bottomed logic we've come to expect from self-help – the same must be true of all big changes. And therefore it must take 21 days to change a habit, maybe, perhaps!

  • 2
    Interestingly enough though, reading through the journal article it looks like there might be a grain of truth to the 21-day habit claim. Based upon the charts it looks like habits tend to be formed after 20 to 40 days and it looks like if you don't have the habit by the end of that period you are unlikely to acquire it. I'd be curious to see what the median time is for 50% of the population to acquire a habit is. Also, what do you mean by "pseudo-scientific" in this case?
    – rjzii
    Mar 25 '13 at 12:40
  • @Rob Z: So there is a grain of truth in the idea that if you repeat something for long enough, it is likely to gain "automaticity", i.e. becomes a habit? Sure. I think Maltz's claim is pseudoscientific, in that it is dressed as a scientific claim (the original quoted texts said 'research shows' - the appearance of scientific support helped the meme spread) but there was almost zero actual scientific evidence backing it.
    – Oddthinking
    Mar 26 '13 at 22:45
  • 1
    No, I mean the grain of truth is that 21 days might be a reasonable minimum estimate for how long it takes for something to become a habit. Granted the way it is written is extremely poor but as a potential grain a truth behind the myth...
    – rjzii
    Mar 26 '13 at 23:02

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .