How and when do habits form?

When behaviours are repeated in consistent settings (places, times, with certain people etc..) they become automatic responses, that is, they become habits.

So how do habits form? How long do they take to form? How is this different with each person? And how important is reward (or reinforcement) in this process?

Rewards: do they matter?

It seems unclear how important rewards are.  The behaviourists (see footnote) think that repetition will only occur (with any behaviours) if there is some reward.  They feel that without reinforcement there is no habit.  However, others argue that perhaps the behaviour itself IS the reward, especially in the instance of desired habits.

So how DO habits form?

Phillipa Lally and her colleagues (2010) had a look at exactly this.  They asked 96 students (in truth they paid them 30 quid) to choose a new plus healthy behaviour (eating, drinking or excise) and do it for 12 weeks.

Importantly they asked them to do the habit in response to a cue, and not a set time of day.  So it might be: ‘drink water with lunch’ or ‘go for a run when I get up’.

What did they learn?  If you can find your way through the statistics (non-linear regression, asymptotic curves, and more) they found that mostly our habit formation is this sort of shape:

i.e. an asymptotic sort of shape.  What does that mean? It means – there is some sort of ‘point of diminishing returns’ i.e., a point where we are behaving really habity-like.  Although, after that key point we get ever closer to some habit nirvana; the speed of approach to perfect habity-ness gets ever slower.  We have reached some sort of ‘tipping point’ where the habit is embedding enough in our daily lives (just at the hump of the curve).  Obviously we want to reach this ‘hump’ as soon as we can, at least for some habits.  For other habits we want to get far, far away from said hump.

How are we all different?

The answer to this question i.e, how different are we from each other in habit formation? … seems to be: MASSIVELY different!

Lally and colleagues found the time to reach the ‘hump’ on the curve (or put in scientists speak: the time it took participants to reach 95% of their asymptote of automaticity) ranged from 18 days to a whopping 254 days!

The question on my mind, and maybe yours is; how to reach it in 18 days rather than 254 days?  This was unfortunately not within the scope of Lally’s research but it is certainly on my agenda now.

I have a horrible feeling that it might be something to do with core personality or genetics.

Which makes me wonder about the differences between good and bad habit formation, or habit formation aligned with goals versus not aligned with goals.  Which in itself could be confusing as a bad habit might be goal aligned e.g. a teenager trying to impress an older, inappropriate love interest.

There is good news – taking a break won’t kill it or you.

The good news is, missing a day in your new habit isn’t that serious to your habit formation.  Missing one day did make the next day habit-y-ness a little less strong, but not in a way that was statistically significant.  Which is just scientist speak for not enough to make any sort of meaningful difference.  And most crucially it did not affect the path of the graph.

How many missed days it takes to make a difference (i.e. break the habit-forming path) was not looked at.

Take outs

It is hard to know what I will take away from this one, not knowing my own story i.e how would I compare to the data here? What is my number of days to each my 95% asymptote?

Perhaps I will plan for the worst, put very strong systems’ in place until much, much more than 18 days after starting a new habit.

Footnotes and references

This post is based on the below article. It is part of a series on habits. Select habits from the drop-down on your right to see other awesome posts on the topic (modesty will be the next series).

Behaviourism is a type of psychology focused on observable actions, what we say or do or don’t say or don’t do, rather than focusing on emotions and stuff.  Clearly there are some criticisms of behaviourism, but it has still taught us a lot and is interesting from the perspective of habits, which are behaviours really huh?

Lally, P., Van Jaarsveld, C. H., Potts, H. W., & Wardle, J. (2010). How are habits formed: Modelling habit formation in the real world. European journal of social psychology, 40(6), 998-1009.

By Rachel adapted from

Leave a Reply