True Habits

Link post

Photo by My Name on Unsplash

Habits have been extensively written about in pop-literature for the last couple of years. Many claims have been made!

Want to get an awesome body? Habitify it!

Want to become great at your studies? Place gummy-bears on each page to train your pesky mind.

Want to get rid of stress? Sit in the same spot and breathe through your nose for 5 minutes, and to ensure you keep doing it, track it in a spreadsheet.

But how well does this fit with the habit literature? And why don’t more people train themselves with gummy-bears? Let’s find out!

A note on “why now”

Many of us are in unfamiliar contexts, being forced to work from home to reduce contagion.

This presents to us both a challenge and an opportunity.

We’re forced to work in unfamiliar contexts and, if you’re anything like me, this means that your work time is now being intruded upon by a bunch of thoughts relating to habits that are great when you relax – but gaming isn’t going to get your thesis done.

Luckily, habits are more easily changed when contexts change. If you want to get new, appropriate habits at home, or get rid of something you’ve been doing, the time is now!

(A short side-note: This will not be a review of all habit-literature, but rather small dives into results I’ve encountered that were interesting and/​or ran counter to what I expected).

What defines a true habit?

If you’re like me, you’re slightly provoked by the title. “True habits” you say? What are “false” habits?

I used to think that doing the same task repetitively would be sufficient to form habits, or was in fact, the entire definition of a habit.

To question that assumption, we must first ask ourselves, “why do we prefer habits to to-do lists?”. If we can just do what we want, when we want to, without getting tired, why bother with this habit business at all?

The answer lies in automaticity. We want to be able to perform our morning routine without thinking too much about it, while tired, stressed and under time-pressure.

It turns out that some actions that change behavior are detrimental to forming automaticity. This is (obviously) really important, and runs counter to much of what’s written in the pop-psych litterature.

But first, what is automaticity?

In the habit literature, it’s operationalised as a high score on the Self-Report Behavioural Automaticity Index (SRBAI) (Gardner 2012). Once again, science excels at acronyms that are both impossible to pronounce and remember.

Luckily, the questionnaire is simple. Participants are asked to rate how much they agree with the statement: “Behavior X is something I”:

  1. do automatically

  2. do without having to consciously remember

  3. do without thinking

  4. start doing before I realize I’m doing it

Impressively, this explains a large amount of the variance in whether the behavior is executed (r+ = .47, 95% CI .39; .43), (Gardner 2012).

How do habits work?

Habits work through a simple loop, Cue->Action->Feedback. You see popcorn (cue), you buy and eat popcorn (action), you taste the salty, buttery goodness of popcorn (feedback) (Duhigg 2012).

The cue initiates the action which is reinforced or weakened by the reward.

How come, then, that we don’t masturbate furiously when we see a sexy person on the street? After all, sexy features are a typical cue, and masturbation feels good (feedback). Few of us have tried it, so we haven’t gotten any negative feedback either.

We’ll turn to this question, and how to use the answer(s) to our advantage, in the next section.

Taking charge: Intentional (de)training

Before undertaking any endeavour, we must consider the costs and the benefits. Changing habits, especially ones that don’t fit our goals, can obviously come with great benefits – but how much effort must we invest?

You may have heard that “it takes 66 days to form a habit”. Like any fact without confidence intervals or error bars, being skeptical of this number is good practice. Where does it come from?

Turns out this is all from one study, Lally 2009. They had 39 participants train a variety of habits, like doing 50 sit-ups after their morning coffee or going for a 10 minute walk after breakfast. Once each day, each participant filled out the Self-Report Behavioural Automaticity Index (SRBAI).

Fit a quadratic regression, measure time to 95% of asymptote (ie. when the curve flattens out) and hey presto, you’ve got yourself a pop-psych number!

66 is a pretty number, I’ll admit. And I think the methodology of the study is sound, if the Self-Report Behavioural Automaticity Index (SRBAI) can be used sequentially. The authors acknowledge these limitations.

But what is often forgotten is the variance. The 25% and 75% percentile are at 39 and 102 days. This means that only 50% of participants reached 95% of “peak automaticity” somewhere between 39 days and 102 days after starting the habits. 25% took a shorter amount of time, 25% took a longer amount of time.

What can we take away from this? Habits are formed over weeks or months, so expect to wait some time before you can do them without paying attention. And don’t expect there to be a magic threshold; expect steady improvement.

The effort is often worth it, though.

Bad habits and good habits

The point of mastering your habits is, at least in my view, not to force ourselves to do things we don’t want, or to remove fun and pleasurable experiences. Guilty pleasures are still pleasures after all.

But sometimes the cost of a habit is higher than the benefit. And sometimes a habit is activated in circumstances where it doesn’t bring much benefit at all!

Consider Neal 2011. Participants were sent to a theater to watch a movie and served either fresh or stale popcorn.

To the surprise of no one, participants liked fresh popcorn more than stale.

For participants with low habitual popcorn-eating, the stale popcorn was less attractive, so they ate less of it. But for participants with high habitual popcorn-eating, the taste of the popcorn had no effect on the amount consumed!

This is the first half of the answer to “why don’t we masturbate in the street”. For actions with low habituality or strong intentions, we choose what suits our goals best. Since very few of us enjoy a night in jail, masturbating doesn’t even enter consciousness.

However, gabitual behavior can clearly be maladaptive. How, then, do we take charge?

Breaking conflicting habits

For strong habits, your intention is weak sauce. We saw this already with fresh vs. stale popcorn. What about something stronger than a taste cue, like Implementation Intentions (Trigger-Action-Plans)?

Some success has been noted, but only 1 study was done outside of the lab. Webb 2009 randomised adolescent smokers who wanted to stop smoking to either implementation intentions or a control condition (listing advantages of seatbelt use).

They found a moderate effect of forming implementation intentions, moderated by habit strength.

If intention is weak, what works, then?

1. Remove the cues/​change the context

Context-changes are a great time to change habits! Some habits don’t even generalize across contexts; for the popcorn eating experiment, no difference was found if the movie was shown in a meeting room rather than a theater.

Be it changing travel habits during office relocations (Walker 2015), optimising your commute during worker strikes (Larcom 2017), or a host of other cases (Verplanken 2016, Bamberg 2006, Thøgersen 2012), context-changes are where it’s at!

This is the second half of why we don’t masturbate in the street: It’s an unknown (and inappropriate) context.

How can you implement this in your life? Personally, during the beginning of the COVID-19 quarantine I had tremendous difficulties focusing on work. Every time I was switching between tasks, I was tempted to “just play one game”. I had to fight urges all the time. This led to me getting much less work done.

Armed with the insights above, I added two desks to my room; 1 for work, 1 for leisure. Two days later, my desire for gaming during work was completely gone.

2. Make the habit harder to perform

If you like gaming with your friends, but you find yourself gaming more than you’d like, why not change your password to something randomly generated and give it to your friend?

If you find that visiting reddit or hacker news has become encoded in muscle memory, you can break the cue->action->reward cycle.

You can use the Chrome add-on Leechblock to block websites. Or, if you want something stronger, don’t just block, redirect it to a website that’s aversive to you. I used a political party I’m strongly opposed to. This will break the near-unconscious habit in no time.

> Have you detrained habits? Which benefits did that bring you?

Once again, the goal is not to remove all guilty pleasures. Leechblock allows manual overrides, so I can still access those sites when I choose to deliberately. But the habitual “this task is difficult” (cue) → browse reddit (action) → fun gifs (reward) loop has been broken.

If Leechblock isn’t your cup of tea, you might consider Crackbook or In Motion. For Gmail, I highly recommend Inbox When Ready.

These two rather simple strategies have been tremendously useful for me. Now we turn to how to add new habits to your stack.

Forming endorsed habits

To intentionally form habits, we must change our behavior and/​or our environment to facilitate habit formation.

Use Implementation Intentions as training wheels

Implementation intentions are based on a trigger and an action, eg. “When I sit on the edge of my bed, I will do 10 push-ups”. They are pretty good at changing behavior, too! A 2006 meta-analysis (Gollwitzer 2006) found a medium-large effect size (Cohen’s d = 0.65).

However, as implied in the techniques name, implementation intentions need intention. They’re dependent on you remembering to do the action based on the trigger.

However, if you choose your trigger carefully, implementation intentions can transition into habits.

Which triggers are good, then? Most habit-apps would have you believe that you can rely on their reminders – but is that really a good idea?

Stawarz 2015 had participants send in a text with what they ate after each meal. The trigger group were told to do it right after lunch, and the reminder group received a text at a set interval.

Unsurprisingly, the reminder group were reminded to send the text and had higher adherence.

Error bars are standard-deviations.

However, they also had a lower automaticity score, implying that they performed the action not out of habit, but because they were reminded.

Error bars are standard deviations.

This implies that acting on reminders requires intentionality. That’s all fine and good, but that’s not why we’re trying to habitify our actions.

Instead, set a good trigger. Make it specific (rather “after finishing my plate” than “after dinner”), choose something that’s consistent (prefer “after getting dressed” to “after getting ready for work”) and preferably chain it to something else in your routine.

There is a bunch more good advice out there for implementing habits. I’ve made a short spreadsheet here, with some added motivation from the WOOP-technique. It’s definitely a little rough around the edges, but rather share something that isn’t perfect, right?

Conclusion

I hope you found this interesting! We’ve talked about what separates habits from repeat intentional behavior, how long it takes to make or break a new habit, when habits are strong and weak and how to use these insights to your advantage.

Now is the time to put it to use. Which habits would you like to add? And which habits are standing in your way? They’re in your hands now.

If you have any questions/​comments, I’d love to hear them below.