What a habit actually is

Most of what you will read about habits rests on a definition that isn't the one the research uses, and almost every confusion in the subject comes from that.

So this lesson does one thing. It gets the definition right, shows you what it rules in and out, and says what this course is for. Nothing here will help you build one yet. Lesson 3 does that, and it will not work if you're aiming at the wrong target.

The definition, and it is not the everyday one

In ordinary speech a habit is something you do a lot. Biting your nails, checking your phone, a glass of wine on a Friday.

That is not what the word means in the literature this course is built on. Gardner, Lally and Wardle put the distinction plainly in a piece written for general practitioners: "While often used as a synonym for frequent or customary behaviour in everyday parlance, within psychology, 'habits' are defined as actions that are triggered automatically in response to contextual cues that have been associated with their performance".¹

Their two examples are worth having, because both are things you do without ever having set out to learn them at all. "[A]utomatically washing hands (action) after using the toilet (contextual cue), or putting on a seatbelt (action) after getting into the car (contextual cue)".¹

Read the definition slowly, because it has three parts and each one does work.

There is a cue, which is a context. A place, a time of day, an event, the end of another behaviour. There is a response, which is the action. And there is an association between them, which is the thing that was built and the thing that does the triggering.

What's missing from that list is you deciding. That absence is the whole point of it.

And the association is built by one thing. The same paper states the mechanism: "Decades of psychological research consistently show that mere repetition of a simple action in a consistent context leads, through associative learning, to the action being activated upon subsequent exposure to those contextual cues".¹

Three words in that sentence do the work. Repetition, because one pairing builds next to nothing. Consistent context, because the association is to a context, so a shifting one gives the learning nothing to attach to. And associative learning, which is the ordinary mechanism by which two things that keep occurring together become linked, and which does not require you to intend any of it.

Frequency is not automaticity

Automaticity is the word this literature uses for the thing the definition turns on: a behaviour starting on its own, from the cue, without being decided about. It is what the third part of the definition names.

The test that separates the research sense from the everyday one is short, and it is the single most useful thing in this lesson.

Two constructed people both go running three mornings a week, because no source read for this course describes an individual's habits.⁵

The first thinks about it each morning. Some mornings she talks herself into it, some mornings she does not, and over a year it comes out at about three. The second's usually laced up and out of the door before the question of whether to go has properly formed.

Same behaviour. Same frequency. Only the second one is a habit in the sense that makes any of this research apply, because only for the second one has the decision dropped out.

What a habit buys you

Two things, and the first is much larger than the second.

A habit doesn't depend on your wanting to do it. The same paper puts the mechanism in one sentence: "Once initiation of the action is 'transferred' to external cues, dependence on conscious attention or motivational processes is reduced. Therefore habits are likely to persist even after conscious motivation or interest dissipates."¹

That is the offer. Not that you will want things more, or want them more consistently, but that for one specific behaviour the wanting stops being load-bearing.

And a habit is cheap to run. In two hourly-diary studies of students, published in 2002, Wood, Quinn and Kashy counted a behaviour as habitual when it had been "performed almost daily in stable contexts", which is a criterion of frequency and context rather than the automaticity test above, which isn't the same thing. Lesson 2 is about that difference and why it matters. Within it, they found that when people were doing something habitual they "were likely to think about issues unrelated to their behavior, presumably because they did not have to consciously guide their actions", and reported less stress than during non-habitual behaviour.² Lesson 2 takes that study properly.

Set those two beside what you already know, because this course starts one step further on than the last one did. Time Management lesson 4 gave you if-then plans and was careful about their limit: an if-then plan is a bridge from wanting to doing, and it does nothing at all about whether you want the thing. Everything in that lesson assumed a goal you actually hold.

A habit is the case where that assumption stops mattering, because the behaviour is no longer being fetched from the goal each time. That's what this course adds, and it is also why it can't help you want things.

What the definition does not say

Three things, and readers get caught by all three.

It says nothing about the habit being a good one. The definition is a mechanism, and mechanisms are indifferent. A useful case, and an uncomfortable one: after the 2007 smoking ban in United Kingdom public houses, many smokers reported "finding themselves lighting a cigarette as an unintentional habitual response to drinking alcohol".³ That is quoted from a 2022 methods paper this course has read, which is reporting other researchers' work rather than its own, and this course has not gone back to the study behind it. The drinking was the cue. The lighting was the response. The association had been built by repetition and was not consulted about the ban.

And it says nothing about discipline. This is the one worth pausing on, because it is what most readers arrive holding. A habit isn't a thing disciplined people have and undisciplined people don't. It is an association, built by repetition in a context, and the mechanism builds them in everybody regardless of what they think of themselves. Lesson 3 is the evidence for that.

It says nothing about the habit being chosen. Nobody sets out to learn to check their phone when they sit down; it is not a plan anybody makes. The association gets built anyway, because the mechanism does not require your consent, only your repetition.

And it says nothing about how you feel while doing it. A habit can be enjoyable, tedious, or unnoticed. What it can't be is deliberated about each time, because then it is something else.

Three behaviours, sorted

All three are constructed, like the runners above.⁵ Take them one at a time and test each against the three parts: a cue, a response, an association doing the triggering.

One. A man puts his keys in the same bowl every time he comes in. He hasn't thought about it in years and notices only when they are not there. Cue, coming through the door. Response, keys in the bowl. And the association is doing the work, because he did not decide today. A habit.

Two. A woman does the household accounts on the first Saturday of the month. She has done this for six years without missing one. She puts it in the calendar, she is glad when it is done, and on the morning itself she gets herself to the desk. Cue, arguably the first Saturday. Response, the accounts. But the association isn't what gets her there, and she knows it, because when the first Saturday is awkward she moves it and does it anyway. Not a habit, and she is more reliable than most people's habits. Reliability and automaticity are different achievements.

Three. A teenager opens a particular app whenever he unlocks his phone. He does it while waiting for a kettle, between sentences, on the stairs. He often could not tell you he'd done it. Cue, unlocking. Response, the app. Association, plainly. A habit, and one nobody chose.

Notice that the sorting has nothing to do with which of the three you would want. Case two is the admirable one and it is the one that fails the definition.

What this course promises

It won't make you more disciplined, and that is a statement about what the research is rather than modesty.

What the evidence in this course describes is a mechanism by which a behaviour stops needing a decision. That is a real thing, it is measurable, and lesson 4 gives you the measurement. It isn't the same as a person becoming someone who does hard things, and a course promising the second on the strength of evidence for the first would be doing the thing this course exists to teach you to spot, and this course has not opened a single one of the popular books on the subject to check whether they do it.

There is a second half to the course and it is less comfortable. Lessons 5 and 6 are about self-control, which is what happens when a decision is needed, and the famous account of it lost most of the evidence that supported it, while what is proposed in its place is less tidy than either the model or its debunking. The arc there is worth knowing before you start: an elegant model, a 2010 meta-analysis of 198 published experiments behind it, and then a preregistered replication across 23 laboratories in 2016 whose pooled estimate could not be told apart from zero[6]. The course tells that story straight, including what the replication's authors say their own result can't settle.

And one question is yours and the course will not answer it. Whether you should want your life run by habit at all.

The case for is what this lesson has been describing, which is not the same as the lesson having argued for it: a habit is behaviour that survives your worst weeks, and a life with more of them in it spends less of itself on getting started. The case against is that a habit is by definition something you do without deciding to, that the mechanism is indifferent to whether the thing is worth doing, and that a person can automate their way into a life they never chose one cue at a time. Both of those are true at once, which is what makes it a question rather than a problem to be solved.

This course has not the standing to settle it, and lesson 7 comes back to it with the whole course behind you. What the course offers meanwhile is narrower, and it's this: what a habit is, what it costs to build one, and what the evidence does and does not show.

A gap in the evidence, named now rather than discovered later

Nearly every participant in this literature is a student or a self-selected volunteer. The diary studies in lesson 2 are students. The field experiment in lesson 3 is a campus cinema. The replications in lesson 5 are undergraduates. The study in lesson 4 is the broadest of them, and it is 96 self-selected volunteers, followed for twelve weeks, published in 2010. This course has not been able to open that paper, which is behind a paywall, so everything it is cited for here and in lesson 4 comes at one remove through two papers by one of its own authors.⁴

And a large share of the habit literature comes from two research groups. That is not a criticism of either, and it does mean the field is smaller than the volume of writing about it suggests.

Where a later lesson reasons past the evidence, it'll say that it's reasoning.

Three things left out on purpose

Breaking a habit. The literature on breaking is thinner than the literature on forming, and what exists is mostly about changing the context. Lesson 3 says what little can be said honestly. A course that handed you a breaking method would be inventing one.

Addiction. A different literature, a different mechanism at the edges, and real harm available if a course treats them as the same thing. This course does not.

The neuroscience. There's a real account involving the basal ganglia and this course hasn't read it. A lesson of brain diagrams would be decoration, and the honest thing's to say so.

Practice

Connections

Back. This is the first lesson, so nothing here depends on the course. It leans on three earlier courses on the Core and re-teaches none of them. Time Management lesson 4 taught you if-then plans and named their limit, which is the doorway this course comes through. Logic and Argument lesson 2 taught you to test a definition by what it rules in and out rather than by whether it sounds right, which is the whole of this lesson. Using AI Effectively lesson 11 taught you to ask what a claim was measured on and when, which is why the gap in this literature is named above rather than in a footnote.

Forward. Lesson 2 measures how much of an ordinary day meets this definition, and traces the figure everybody quotes for it. Lesson 3 is what the cue is doing, and it is the first lesson with a technique in it. Lesson 4 is how long a habit takes and what happens when you miss a day. Lessons 5 and 6 are self-control. Lesson 7 comes back to the question this lesson declined.

Go deeper

• Making health habitual: the psychology of 'habit-formation' and general practice (British Journal of General Practice, 2012). Three pages, free, and this course has read all of it. Written for doctors rather than for readers of self-help, which is why it is the clearest short statement of this subject in print. It ends with a tick sheet a GP can hand to a patient, and lesson 4 uses it.
• Habits in Everyday Life (Journal of Personality and Social Psychology, 2002), posted by one of its authors. This course has read the abstract and the diary-study results and no further. The results are where lesson 2 comes from. The method is where the definition of one behaviour is settled, and this course has not read it.

Sources

1. Benjamin Gardner, Phillippa Lally and Jane Wardle, "Making health habitual: the psychology of 'habit-formation' and general practice", British Journal of General Practice 62(605), December 2012, pages 664 to 666. Read in full, open access. Supports: the quoted definition and both of its examples; the quoted sentence about initiation transferring to external cues and habits persisting after motivation dissipates. Two of its three authors are authors of the study lesson 4 rests on.
2. Wendy Wood, Jeffrey M. Quinn and Deborah A. Kashy, "Habits in Everyday Life: Thought, Emotion, and Action", Journal of Personality and Social Psychology 83(6), 2002, pages 1281 to 1297. Read in part: the abstract and the diary-study results. Supports the quoted finding about thoughts unrelated to the behaviour and the lower reported stress. Scope: two hourly-diary studies of students.
3. Benjamin Gardner, Amanda L. Rebar and Phillippa Lally, "How does habit form? Guidelines for tracking real-world habit formation", Cogent Psychology 9:1, 2022, article 2041277. Read in substantial part, open access. Supports the smoking-ban case in that paper's own words, which is what the body quotes. That paper is describing other researchers' work rather than its own, and this course has not gone back to the study behind it.
4. Phillippa Lally, Cornelia van Jaarsveld, Henry Potts and Jane Wardle, "How are habits formed: Modelling habit formation in the real world", European Journal of Social Psychology 40(6), 2010, pages 998 to 1009. Paywalled, and this course has not opened it. Everything it is cited for is reported at one remove through sources 1 and 3, both of which have Lally as an author, and lesson 4 says so where its figures appear. Cited here only for the sample of 96 volunteers.
5. Every person in this lesson is constructed, and so is every detail attached to them: the two runners, the three sorted behaviours, the swimmer and the phone-checker in the quiz, and the examples in the predict block. No source read for this course describes an individual's habits. They are built to show a distinction the research does draw. The claim that the association in the smoking case had been built by repetition is an inference from the mechanism rather than something the source states.
6. Mark S. Hagger and others, "A Multilab Preregistered Replication of the Ego-Depletion Effect", Perspectives on Psychological Science 11(4), 2016, pages 546 to 573, and the 2010 meta-analysis it followed, Hagger, Wood, Stiff and Chatzisarantis. The replication's abstract and limitations section were read verbatim; the 2010 meta-analysis has not been opened and its count of 198 published experiments is taken from the 2015 re-analysis that quotes it. Cited here only for the two figures in the preview above, and taken properly in lesson 5.