Join 3,523 readers in helping fund MetaFilter (Hide)


Not 21 days
April 19, 2014 1:40 PM   Subscribe

How Long does it actually take to form a habit? Answer: Not 21 days. ...Maltz's work influenced nearly every major "self-help" professional from Zig Ziglar to Brian Tracy to Tony Robbins. And as more people recited Maltz's story -- like a very long game of "Telephone" -- people began to forget that he said "a minimum of about 21 days" and shortened it to: "It takes 21 days to form a new habit." A study debunks a popular self-help myth.
posted by storybored (19 comments total) 20 users marked this as a favorite

 
But...but...if you can't trust self-help books to be scientifically accurate, who can you trust?
posted by escape from the potato planet at 2:19 PM on April 19 [3 favorites]


The entire self-made, pep-talk, self-empowerment industry is built on bullshit and placebos, pitched by creeps and narcissists.

But it makes people try something new in an attempt to break out of their ruts. Not that it often succeeds.

The only real problem is when people become addicted to experiencing the group experience, and end up spending small fortunes on seminars, and going nowhere in real life.
posted by five fresh fish at 2:53 PM on April 19 [7 favorites]


James Clear
Entrepreneur, weightlifter, and travel photographer


Broscience.
posted by charlie don't surf at 3:17 PM on April 19 [8 favorites]


I hate when I want to know the answer to a meta thread, but nobody has written it yet. So if you don't want to read the whole damn thing to just know the answer:

In other words, if you want to set your expectations appropriately, the truth is that it will probably take you anywhere from two months to eight months to build a new behavior into your life -- not 21 days.
posted by hal_c_on at 4:20 PM on April 19 [24 favorites]


I heard it was 14 days! So now I can build new habits and change my life faster than the rest of you who believe it takes 21 to 240 days. Hah!

and going nowhere in real life

I don't know, my friend moved from the frozen north to southern California to hang out with T. Harv something-or-other, which seems like a pretty good gig. (Well, except for the loss of your immortal soul, but people don't seem to worry so much about that stuff any more.)
posted by sneebler at 5:02 PM on April 19


People seem to drop out of diet and exercise regimes pretty quickly.
posted by IndigoJones at 5:27 PM on April 19


On average, it takes more than two months before a new behavior becomes automatic -- 66 days to be exact.

Good odds I can change my life significantly in two months? Shorter than I would have thought. I take that to be really good news.
posted by justsomebodythatyouusedtoknow at 6:32 PM on April 19 [2 favorites]


The abstract for the research reported in the linked article says The time it took participants to reach 95% of their asymptote of automaticity ranged from 18 to 254 days. Is it possible to explain what that means to somone who is statistically illiterate. The linked article would suggest that it means it takes 8 months to form a habit, but I don't know how that translates.
posted by layceepee at 7:23 PM on April 19


Um, what's a habit? They mention running. I've been jogging for years and years and years, and it's a slog every single time - nothing habitual about it that I can tell. I don't like to exercise and do it only because of health reasons. But I dread it Each. And. Every. Time. It hasn't gotten better since I started decades ago. I strongly doubt that after 25 years, one day I'll suddenly achieve "habit" status when it comes to jogging. I do it deliberately, not out of habit, and I have to compel myself every single time. Nothing automatic or habitual about it.
posted by VikingSword at 8:44 PM on April 19 [7 favorites]


VikingSword, I think that there are quite a few people who would love to get into your habit of jogging that would disagree with you. A habit doesn't have to be enjoyable. Maybe your habit is to "deliberately" and consistently tell yourself to go jogging.
posted by eye of newt at 9:04 PM on April 19 [3 favorites]


James Clear
Entrepreneur, weightlifter, and travel photographer

Broscience.


Whoa, I went to college with him. To be fair, if you click through to his website, it appears that each of those titles is true - and not in the "I-took-photos-of-my-vacation-therefore-I-am-a-travel-photographer" sense. (Not claiming to be a fan of his project, though.)
posted by Austenite at 9:32 PM on April 19


You hear that, Sams Publishing? You'd better drop those deceptive titles soon!
posted by b1tr0t at 9:40 PM on April 19


I guess that explains why I did 100+ days of meditating and still never turned it into a habit I enjoyed sticking with.
posted by jenfullmoon at 9:53 PM on April 19


I definitely think that enjoying something and it being habit are way different. I cannot run far and I cannot run fast, but I am tickled pink at this point that I can run at all, and I'm trying to go about three times a week to the gym for that purpose. It's basically week 1 of couch to 5k and I may be there awhile, but I push hard every time and every time I feel amazing when I leave. Wheezy, but amazing. I'm very subject to that cardio euphoria thing and my mental health is markedly better for at least the following few hours, and at least a little better overall. But it's not habit. I have trouble going at anything close to the same time each time, I forget to go even when I have time and would feel better for it, at least once a week I get halfway there and realize I don't have something vital I need for the enterprise.

Nobody--or at least very few people--really loves brushing their teeth, but either you have the habit or your don't. I forget that a lot, too, to be honest. I am gradually coming to the idea that I am habit-resistant. It's been good for helping me avoid some addictions that other members of my family are prone to, but bad for establishing healthy routines. And dear god I want to stab people, especially family members, who trot out the "anything you do for 21/30/90 days is a habit you'll keep forever" stuff, because it comes with a serious implication that I couldn't stick with something for even that long and clearly this means I lack Grit. I am glad to have ammunition to throw back. I wish it came with any ideas for how to shorten that time or make up for it.
posted by Sequence at 2:03 AM on April 20 [2 favorites]


The abstract for the research reported in the linked article says The time it took participants to reach 95% of their asymptote of automaticity ranged from 18 to 254 days. Is it possible to explain what that means to somone who is statistically illiterate. The linked article would suggest that it means it takes 8 months to form a habit, but I don't know how that translates.

The authors modelled the increase in automaticity as looking something like this graph. Quoth the article:
Hull's early work (1943, 1951) suggested that the relationship between repetition and habit strength follows an asymptotic curve in which automaticity increases steadily—but by a smaller amount with each repetition—until it reaches an asymptote (plateau).
Their language here is quite sloppy (from my point of view as a mathematician, at least), but hopefully you can see what they're trying to say by comparing their description to the graph I linked.

One key technical point is that the type of curve they used in their model gets shallower and shallower, but never actually becomes horizontal — it's always increasing, though (as they say) by less and less as you move to the right on the graph. So it's no good to ask, "When does automaticity reach its maximum?", because the answer in the model is, never. Instead, they make do by asking, "When does automaticity reach 95% of its maximum?" Quoth the article:
Modelling the habit formation process in individuals allowed us to calculate the time taken for automaticity scores to plateau; operationalized here as time to reach 95% of asymptote. [...] The essential nature of an asymptote—that it is never reached—means it is impossible to assess the time to reach 100% of asymptote, and anything above 95% leads to a very large estimate of time taken because of the shape of the curve. From visual inspection, the points identified using 95% are approximately where the curves appear to reach asymptote, whereas 90% identifies a level clearly below where the curves level off.
(In the graph I linked, the "maximum" is 1, the curve passes 0.95 a little to the left of x=3, and it passes 0.90 at about x=2.3.)

The study had 96 participants, of which 39 had automaticity scores over time that could, in the authors' opinion, be satisfactorily modelled with their chosen class of curves. Quoth the article:
Ten of these participants were performing eating behaviours, 15 drinking behaviours and 13 exercise behaviours. The median time to reach 95% of asymptote was 66 days, with a range from 18 to 254 days.
"Median" means that half the participants took less than 66 days and half took more (or would have, if they continued to behave as modelled). So 66 days is in some sense a typical result (if you're among the 39 people whose automaticity increased in the way the authors' model predicts).
posted by stebulus at 6:53 AM on April 20


Whoa, I went to college with him.

Bro.

I sometimes see broscience like this from weightlifting communities. They try to extrapolate some universal human truth from some guy's exercise theory. I don't think Maltz was necessarily talking about how many sets and reps, or miles run, when he was talking about forming positive habits.
posted by charlie don't surf at 8:01 AM on April 20


Are they just talking about good habits?
It feels like I can for a new bad habit in days, if not hours.
posted by MtDewd at 10:29 AM on April 20 [3 favorites]


I don't think Maltz was necessarily talking about how many sets and reps, or miles run, when he was talking about forming positive habits.

Well, yeah. That's probably why he (Clear) explains the context in which Maltz came up with the "21 days" idea and then focuses on the McLally experiment, which set out to measure health-related behaviors, in particular.

They try to extrapolate some universal human truth from some guy's exercise theory.

This is taken from the HuffPo "Healthy Living" pages. It's billing itself as "exercise theory." If folks read that as a source of universally applicable truths, that's on them; I don't think the article (or the editors who placed it there) are making grander claims than that.
posted by Austenite at 11:26 AM on April 20


In 12 step programs, since they recommend 90 meetings in 90 days, the claim is that it takes 90 days to make a habit.

But the reality— in addictions, even in terms of physiological addictions to substances like opioids— is that it is highly individually variable. One person might take heroin daily for 2 weeks and then get sick if he runs out; another might take for a month and have no withdrawal at that point. One person might take every day for pain and hate every minute of the experience; while another loves it: something like 15% of people actually don't like the high, another 15% slightly dislike, another 30% find it mixed and the rest like it with varying degrees of intensity. about half of those who really like it say "yikes" and stop immediately; the other half are the people we call addicted.

So, this is why anyone who makes universalizing statements about habits or addictions is generally wrong. YMMV is typically the best you can do ;-)
posted by Maias at 3:46 PM on April 20 [1 favorite]


« Older Roxane Gay on The Biggest Loser:...  |  Sometimes, famous landmarks lo... Newer »


This thread has been archived and is closed to new comments