Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Common failure modes in habit formation

14 Post author: RomeoStevens 28 June 2013 05:28AM

In one project, 256 members of a health-insurance plan were invited to classes stressing the importance of exercise. Half the participants received an extra lesson on the theories of habit formation (the structure of the habit loop) and were asked to identify cues and rewards that might help them develop exercise routines.

The results were dramatic. Over the next four months, those participants who deliberately identified cues and rewards spent twice as much time exercising as their peers. Other studies have yielded similar results.

-"Lifestyle Intervention by Self-Regulation of Action (LISA)" study by Stadler, Oettinger and Gollwitzer 2005.

I don't think this topic needs a huge introduction.  Most of us have tried, at some point, to establish a new routine only to have it crash and burn. We came up with and discussed some of the more obvious failure modes at last week's southbay meetup, which generated the material here.  It would be awesome to further refine this.  Particularly, some overarching ontology of failure modes would be useful for turning them into a more mentally compact checklist. So feedback on how this material can be organized and presented better is most welcome.

Failure is Always Failure

"I would have succeeded if it weren't for those meddling kids!" The "perfect plan" that you can't actually execute on is not the perfect plan. Take responsibility for the failure and figure out what's really going on. 

Mental cue: Bad news is good news.

Negative Reinforcement

Taking responsibility for failure doesn't mean beating yourself up over it. If you have bad feelings every time you think about habit X due to past failures you are only reinforcing the act of not thinking about habit X.  Failure means you are aware that something went wrong, which means you can improve.

Mental cue: The process failed, so fix the process. Failure and iteration is part of good processes.


That a good process will yield good results doesn't mean we should fall prey to paralysis by analysis. It also doesn't mean we should give up and go back to the drawing board every time we experience a bump in the road.  People commonly engage in visualizing a perfect version of themselves, who obviously wouldn't have failed.  This is frustrating, demotivating, and possibly what is going on with the planning fallacy. Notice when you are constructing a fictional narrative about how well it is possible to do.  How well would you expect a friend in the same situation to do?

mental cue: The perfect is the enemy of the good. You are your own worst critic.

Going too Big too Fast

In the perfect world of our minds, we choose big, exciting-sounding goals and execute on them flawlessly. We become fit, write the next Pulitzer-winning novel, and found a successful startup.  We usually gloss over the fact that getting fit actually means doing pushups, writing a novel involves writing individual pages, and running a successful startup involves emptying your own wastepaper basket. When there is a disconnect between our big goals and everyday actions we don't feel motivated to do those mundane tasks.  Goal factoring, and other techniques for connecting our little goals to our big goals help here.

Mental cue: Granularize

Assuming Constant Motivation

When we create sub-goals we choose things we think we can do. "Of course I can walk 30 minutes everyday." We ignore that when we are creating and evaluating plans we are likely to be in a highly motivated mood. Of course everything seems easy when we are in a motivated mood. Apportion your limited budget of highly motivated time to ensuring that you will be surrounded by cues that encourage your new habit, whether this be people, things, or situations.  This can be as simple as "surrounding yourself" with alarm apps that cue you to do the things you precommitted to doing.

Mental cue: You are the average of your surroundings.

Not Quantifying the Results

Far goals are often qualitative.  We're not sure how much we want to improve by, we just know it's a lot. The problem is that qualitative goals aren't very motivating in terms of actual actions. "I want to get better about responding to emails." Notice the word "better". Contrast with "I want to cut the number of emails I don't respond to by 50% over the next 2 weeks." Now we're getting somewhere, and we have somewhere to start. This is also related to the concept that motivation is hard to maintain when one of our sub-agents has an objection to what we're doing (usually because they aren't convinced it is a good use of time.)

Mental cue: Be specific.

Brittle Plans

This bit was somewhat disorganized. But it involves having a Plan B, as well as figuring out when you are going to reevaluate and update your plan. Also recognizing that what matters in habit formation is getting it mostly right and one shouldn't give up just because they screwed up one time, or even several times.

I'm all fired up to form new habits, now what?

If you don't have anything you're currently working on I suggest instilling the habit of researching new, possibly beneficial habits to have.


Note: In writing this I'm noticing similarity to SMART goals.  Perhaps adapting that would be better since it's already nice and memorable.

Comments (19)

Comment author: Ratcourse 28 June 2013 11:14:36AM 10 points [-]

WRT S.M.A.R.T. goals, Nick Winter says in the motivation hacker:

When you do pick your goals, forget the advice about SMART goals. Use Piers Steel’s slightly improved CSI Approach. Your goals should be Challenging (if they’re not exciting, they won’t provide Value); Specific (abstract goals can leave you vulnerable to Impulsiveness, since it’s not clear what you need to do); Immediate (avoid long-Delayed goals in favor of ones you can start now and finish soon), and Approach-oriented. (As opposed to avoidance goals, where you try not to do something, you should instead reframe it positively as an attempt to do something—it just feels better.)

Nick Winter knows about habit formation

Comment author: amitpamin 30 June 2013 04:04:40PM 1 point [-]

I agree with the challenging bit, but for a different reason. Quoting from Piers Steel, "We are motivational misers who constantly fine-tune our effort levels so that we strive just enough for success."

For low complexity goals, up to a point, there is a linear relationship between goal difficulty and goal performance, even when the reward is held constant. That is, more difficult goals require more motivation; provided that the goal is valuable, that motivation is provided.

The difficulty with choosing challenging goals is ensuring that you feel motivated, not excited. Excited is thinking of the benefits and feeling positive emotion in anticipation. Motivation is the energy necessary to actually complete all the crap in between, like doing your pushups. I use mental contrasting for this. It works better than nothing, but still leaves much to be desired.

Comment author: buybuydandavis 28 June 2013 09:12:31AM *  8 points [-]

If you have bad feelings every time you think about habit X due to past failures you are only reinforcing the act of not thinking about habit X.

Ding! Ding! Ding! We have a winner!

It took me a long time to figure that out. If you feel bad when you're trying to solve your problems, you are going to avoid solving them.

And thank you, I was just thinking about habit formation with respect to personal character development as George Washington and Thomas Jefferson would have it. Why don't I do useful things as much as I could? Why are there so few people who do? Some of it is bad attitude, but much of it is just bad habits, or other habits.

Comment author: Pablo_Stafforini 28 June 2013 02:37:10PM *  2 points [-]

Ding! Ding! Ding! We have a winner!

Agreed. Incidentally, "Don't beat yourself up" is one of the top-rated pieces of advice in the Boring Advice Repository.

Comment author: Viliam_Bur 29 June 2013 06:18:57PM *  1 point [-]

I used to have "friendly" people around me, who did the job of reminding me of my past failures whenever I thought about something new. For many years it reinforced me against having new plans. But luckily, it also gradually reinforced me against sharing my plans with this type of people. The important thing was to notice that not all people react like this -- because I sometimes need to discuss my plans with someone.

Comment author: gothgirl420666 28 June 2013 11:23:45AM *  7 points [-]

Regarding "Going too Big too Fast", I always feel the need to constantly remind myself that for any given skill I need to go from "bad" to "average" to "good" rather than just from "bad" to "good". This seems to be related to certain psychological theories regarding the ego. For whatever reason, my mind is okay with "I am physically weak, but I will eventually put in effort to remedy this condition", and "I have worked incredibly hard and it has paid off and now I am super-strong", but "I have put in a significant amount of time and effort and now I am about as strong as the average person" seems kind of pathetic.

In writing this I'm noticing similarity to SMART goals. Perhaps adapting that would be better since it's already nice and memorable.

SMART goals are kind of dumb. SMART stands for Specific, Measurable, Attainable, Relevant, and Time-constrained. The problem is that specific, measurable, and time-constrained basically just mean "specific", and attainable and relevant basically just mean "realistic". So it's redundant, but it's also missing two other key characteristics. A better four-point criteria for goals is:

  • Specific - clear success/failure condition
  • Realistic - "write a book", not "write a New York Times bestseller"
  • Challenging - the goal should reflect the upper bound of your capabilities, this is because people rise to meet whatever expectations they give themselves
  • Motivating - you should feel emotions when you think of your goal, and it should be tied to your long-term goals and abstract values

This is all directly out of some book I read, by the way, but I can't remember which one.

Comment author: witzvo 24 October 2013 11:47:39PM *  1 point [-]

... Specific, Realistic, Challenging, Motivating ...

If anyone knows a source for this (or related research), I'd appreciate it.

Comment author: malcolmocean 30 June 2013 12:53:55PM *  5 points [-]

Since people are listing criteria for goals, I'd like to put forth the "steps" that motivational speaker Zig Ziglar describes (which I believe to be related to, although not necessarily based on, the research summarized in the paper Building a Practically Useful Theory of Goal Setting and Task Motivation—A 35-Year Odyssey published in 2002).

Ziglar describes a study (on Harvard grads, I believe) where they asked them about their career goals and then found out how many had done most or all of the following steps (answer: ~20% and ~2% respectively).
1. State what the goal is.
2. Write it down.
3. Write down why you want to achieve it. (I do a brief goal-factoring tree for this)
4. Write down what potential obstacles are in your way ("if you don't have obstacles, you'd already be there")
5. Write down who can help you reach the goal. (people, groups, organizations)
6. Make a list of what you need to know to reach the goal.
7. Develop a plan of action to reach the goal.
8. Set a date at which you plan to have met your goal (or, I suppose, to reevaluate).

In the study, which was longitudinal, they determined that the people who had done all of these steps were way more likely to have succeeded. Personally, I've done this several times and found it very helpful. More on that below.

Random thoughts:
* My apologies for not citing the actual study, but my exposure to it is only through Zig, who refers to it in a recording (I had to transcribe the steps as typed here).
* Obviously in the aforementioned study there could be selection bias for "who cared enough to do these steps" but I'm pretty sure someone has tested it as an intervention as well.
* Any goal for which you can do this is probably also SMART, but the reverse isn't remotely true.
* One of the best parts of these steps is they don't actually require a lot of work. Now, learning what you need to know to reach the goal may require work, but it's work that was on your path anyway. Someone complaining about this might do well to revisit the Litany of Gendlin.

Upon consideration, Step 7 looks like it could be a fair bit of work, but I think even having a basic outline/story there totally counts. At any rate, I used the related Pick Four system for my album-recording project last year, and my album plan was about 120 words long and written without pre-planning, yet it revealed several steps that I had left implicit, which if I hadn't thought about them at the start would have been left too late. One of them was figuring out exactly where I'd be recording. It would have been easy to think I could do that after practising & choosing which songs to record, but it ended up being completely critical to do them in parallel.

Comment author: ChristianKl 12 May 2015 10:31:02PM 0 points [-]
  • My apologies for not citing the actual study, but my exposure to it is only through Zig, who refers to it in a recording (I had to transcribe the steps as typed here).

Motivational speakers quite often refer to a non-existing goal study. http://sidsavara.com/personal-productivity/fact-or-fiction-the-truth-about-the-harvard-written-goal-study

Comment author: malcolmocean 13 May 2015 05:24:02PM 0 points [-]

Baha, that doesn't surprise me at all. It's worth noting, for those who don't click through or who don't read the whole article you linked, that Sid found a study conducted that does support the model that writing down goals makes you more likely to achieve them.

See some of the data for that here: http://cdn5.sidsavara.com/wp-content/uploads/2008/09/researchsummary2.pdf

Comment author: ChristianKl 13 May 2015 06:52:16PM 1 point [-]

Sid found a study conducted that does support the model that writing down goals makes you more likely to achieve them.

To be exact it makes it more likely that you think you achieve your goals.

In general it's not hard to find studies supporting any claim. The important thing is whether the studies actually provide reliable evidence.

Comment author: malcolmocean 12 May 2015 10:10:41PM 0 points [-]

Someone discovered this old comment and asked about the album. It is here: http://maleidoscope.bandcamp.com/

Comment author: Qiaochu_Yuan 28 June 2013 07:30:15AM *  3 points [-]

I found Charles Duhigg's The Power of Habit a useful and enjoyable read on habit building, although it is a little Malcolm Gladwell-esque. Anyone here read it, and if so, what did you think of it?

Comment author: Pablo_Stafforini 28 June 2013 02:41:33PM *  4 points [-]

I skimmed it at a Barnes & Noble bookstore, while travelling in New York. My recollection is that the book was serious and well-written, but that most of the actionable advice was summarized in the Appendix, so the rest could be safely ignored.

Free copies of the book may be found here.

Comment author: Qiaochu_Yuan 28 June 2013 06:35:08PM 3 points [-]

I think that's a fair summary, but reading the rest is still helpful as a way to mentally anchor the actionable advice.

Comment author: RomeoStevens 29 June 2013 12:40:54AM *  1 point [-]

Awesome, summary of summary:

Thanks to scientific research, we now not only know that habits are powerful actors, but also understand how they are created. This process is called the “Habit Loop,”

In basic terms, a Cue is a trigger that sends your brain into automatic mode and tells it which habit to use. A simple example is the act of smelling freshly baked cookies. This aroma may trigger a Routine, which can be physical, emotional, mental, or a combination of the three. In the case of smelling cookies, the routine may be to go buy one. This results in a Reward, which helps your brain figure out if a loop is worth following. In the cookie scenario, the reward is the satisfaction of hunger, whether real or perceived. Of course, this routine can result in weight gain over time and the formation of a “bad” habit. Duhigg points out that there is a golden rule of habit change: You can’t extinguish a bad habit; you can only change it! This explains why bad habits are sometimes very difficult to change, no matter how good our intentions are.

The basil ganglia portion of the brain is older and more primitive than the brain’s outer layers.

the basil ganglia can’t distinguish between good and bad habits.

some habits are so strong that they can cause our brains to hang on to them to the exclusion of common sense (e.g., the gambling addict’s belief that he has a “big win” right around the corner, etc.).

Cravings power the Habit Loop and are what make cues and rewards work. In other words, habits derive their strength from cravings.

there are mechanisms to help us ignore temptations. However, in order to do this, we must recognize which craving is driving our behavior. Once we identify the craving, we can more easily create a new habit.

The premise that it is possible to transform a habit, but not get rid of it completely, is what Duhigg calls the golden rule of habit change.

you must keep the old cue and deliver the old reward while inserting a new routine. (e-cigarettes are a poster child for this process)

keystone habits were leveraged to create small wins in order to transform organizations and individuals.

willpower is an essential keystone habit. willpower is the single most important keystone habit for individual success. He also points out that the common belief that willpower is something you either have or don’t have is false. Instead, willpower is a learnable skill. Once a person learns good willpower habits in one area, these habits spill over into other areas. (practice setting alarms and doing what they tell you when they go off over and over again.)

Duhigg concludes that we must take responsibility for the habits that we can control. By identifying and understanding these habits, we have the freedom (and in some cases, the obligation) to change and remake them.

Example in actual use: The Habit Loop The first step is to identify the routine in the loop. He uses a personal example — his routine was to eat a chocolate chip cookie every afternoon, which caused him to gain weight. In order to change the behavior, he needed to identify what was driving it.

The second step is to experiment and find an alternative reward. The goal is to identify which craving is driving your routine, and find something to do instead. To continue the author’s example, this step would require finding a substitution for eating a cookie. If the author found that his craving was for the energy derived from the cookie, then walking around the block may be a suitable alternative. If the author’s craving was hunger, then eating an apple may be an effective substitute.

Duhigg suggests you look for patterns and take notes. For his own purposes, after replacing his afternoon cookie with a walk, upon returning to his desk he would write down the first three things that came to his mind, such as, relaxed, saw flowers and not hungry. The act of writing forces a momentary awareness of what you’re thinking about or feeling. It also helps you remember what you were thinking at that moment.

Once you’ve identified the routine and determined the right reward substitute, it’s time for the third step: go back and identify the cue. In the case of the cookie, the author determined that the cue wasn’t hunger, but the need for a temporary distraction from his work.

After following these steps, you can come up with a written plan. In Duhigg’s case, it was simple: at 3:30 p.m., he would walk to a friend’s desk and talk for 10 minutes rather than buying a cookie. As he points out, the change didn’t take effect immediately, but, in the long run, it worked because he understood how a habit operates. Once he had that understanding, he gained power over it.

Comment author: RomeoStevens 28 June 2013 09:32:14AM 0 points [-]

From reading reviews it seems the interesting claim of the book is that choosing which habits to build is a lot trickier and more important (due to subtle wide ranging effects) than building a habit, which is asserted as fairly straightforward.

Comment author: firstorderpredicate 28 June 2013 07:47:04AM 0 points [-]

Becoming too task oriented as well is a bit of trap. As you tick off lots of unimportant tasks, you pat yourself on the back on how effective you are. You risk adding more and more tasks without considering how they fit in with your goals. (This might be what unconsciously goes in the mind of students and writers when they spring clean their houses instead of writing or studying.) Having said that, having a plan that you then act on (and not ignore) is better than not accomplishing anything at all.

I'm not sure how easy a "Plan B" is. If you can anticipate the ways in which your plan fails, why not just incorporate them into "Plan A"? Alternatively I would suggest adding a little flexibility. Evaluating the outcome of your plans should help also. But I really like the "Granularize" and "Quantifying Results". I wish you had written this three months ago. It explains why I'm having so much trouble accomplishing a task. (Develop a marketing plan for an app to be specific.)

Additionally, I especially suffered analysis paralysis by chasing perfection. I'd spend hours shifting tasks around and feel guilty that I spent so much time not getting stuff done. It was for this reason that I built an app to automate my planning process. (Shamful Plug . Currently free for all LessWrong readers.) As a time saving device it's been ok, but it's real benefit is in knowing that I've committed to accomplishing certain tasks and being able to refer to the plan when I get behind or ahead. There's also nothing better than being able to add tasks at odd moments when they pop into your head, instead of sitting down later and thinking "what did I have to do again?".

So I'll finished then on an obvious note: Whatever process or tool you use, you need to record it and not rely on your memory. Because you're more likely to only remember the urgent tasks and delay the important ones.

Comment author: RomeoStevens 28 June 2013 09:27:54AM 0 points [-]

Goal factoring is awesome and through the collective efforts of the southbay meetup a summary of its use should go up sometime next week.