Charles Duhigg’s The Power of Habit

The skills necessary to ride a bike are multifaceted, complex and not at all obvious or even easily explicable to the conscious mind. Once you learn, however, you never forget–that is the power of habit. Without the power of habit, we would be lost. Once a routine is programmed into system one (to use Kahneman’s terminology) we can accomplish great skills with astonishing ease. Our conscious mind, our system two, is not nearly fast enough or accurate enough to handle even what seems like a relatively simple task such as hitting a golf ball–which is why sports stars must learn to turn off system two, to practice “the art of not thinking,” in order to succeed.

Habits, however, can easily lead one into error. In the picture at right, which yellow line is longer? System one tells us that the l at the top is longer even though we all know that the lines are the same size. Measure once, measure twice, measure again and again and still the one at top looks longer at first glance. Now consider that this task is simple and system two knows with great certainty and conviction that the lines are the same and yet even so, it takes effort to overcome system one. Is it any wonder that we have much greater difficulty overcoming system one when the task is more complicated and system two less certain?

You never forget how to ride a bike. You also never forget how to eat, drink, or gamble–that is, you never forget the cues and rewards that boot up your behavioral routine, the habit loop. The habit loop is great when we need to reverse out of the driveway in the morning; cue the routine and let the zombie-within take over–we don’t even have to think about it–and we are out of the driveway in a flash. It’s not so great when we don’t want to eat the cookie on the counter–the cookie is seen, the routine is cued and the zombie-within gobbles it up–we don’t even have to think about it–oh sure, sometimes system two protests but heh what’s one cookie? And who is to say, maybe the line at the top is longer, it sure looks that way. Yum.

System two is at a distinct disadvantage and never more so when system one is backed by billions of dollars in advertising and research designed to encourage system one and armor it against the competition, skeptical system two. Yes, a company can make money selling rope to system two, but system one is the big spender.

Habits can never truly be broken but if one can recognize the cues and substitute different rewards to produce new routines, bad habits can be replaced with other, hopefully better habits. It’s habits all the way down but we have some choice about which habits bear the ego.

Charles Duhigg’s The Power of Habit, about which I am riffing off here, is all about habits and how they play out in the lives of people, organizations and cultures. I most enjoyed the opening and closing sections on the psychology of habits which can be read as a kind of user’s manual for managing your system one. The Power of Habit, following the Gladwellian style, also includes sections on the habits of corporations and groups (hello lucrative speaking gigs) some of these lost the main theme for me but the stories about Alcoa, Starbucks and the Civil Rights movement were still very good.

Duhigg is an excellent writer (he is the co-author of the recent investigative article on Apple, manufacturing and China that received so much attention) It will also not have escaped the reader’s attention that if a book about habits isn’t a great read then the author doesn’t know his material. Duhigg knows his material. The Power of Habit was hard to put down.


Riding a bike is a skill. If you do it every day, that's a habit.

What's the difference between a skill, a habit, an addiction, and a conditioned reflex?

Riding a bike is a skill, but habits don't have to be done every day. Habits are done the same way every time you do them.

Skills are conditioned responses that are selected at the appropriate time. The amount of time and level of consciousness is determined both by the speed necessitated by the environment and the habits engaged from stimulus.

A typical driver is skilled at driving. They track many moving objects, judge speed, distance, bearing, and heading, traction. They know from experience the stopping distance of their car at various speeds. If they are trained, they will drive defensively by anticipating what a like-minded driver or pedestrian nearby intends to do or might do. In a rental car, they will have some learning to do. They will also often feel out of place with their own car when they get home.

A racecar driver does all that, but with a lot less conscious thought - reflexive responses in milliseconds. Ironically, a race car driver has fewer variables in play. All the cars are moving in the same direction at similar speeds. There are fewer stops and fewer control measures (no lanes, no lights, no signs - only an occasional flag). The pit crew does a lot of their thinking for them.

Addictions are behaviors that stimulate emotional or physical states that are comfortable and familiar. These are often, but not always, associated with pleasure. The psychochemical reaction of an addictive response removes a degree of volition. Your want becomes a need. Inhibitions and moderators are removed.

Commencing to brush your teeth is a conscious act. The manner in which you brush is likely a thoughtless habit that you replicate every time. Applying knowledge of proper dental hygeine to develop effective brushing habits, and knowing how to modify your habits, say, if your primary hand is broken - that's skill. Catching your toothbrush before it hits the ground after accidentally knocking it over is a reflexive response. Not accidentally knocking your toothbrush over at all is a conditioned response.

Habits free the mind for other tasks, but habits have to be conditioned into an effective pattern to be helpful. The conscious mind needs to be alert to prevent unexpected accidents such as slipping on a wet surface, treading more carefully when you would otherwise carelessly walk or step.

>>> Habit tells us that the line at the top is longer<<<

So, do young kids perform better at this task? Not sure at what age the length discrimination ability kicks in.

I've read that those who live in natural settings (ie primitive societies) rarely see a difference between the lines. I'd suggest that it's not a specific age, but a function of how many straight lines surround us in our daily lives.

That's an astute observation. Our society is engrossed with right angles that condition us to fall for the illusion.

I was thinking of this on a simpler level - we see more two dimensional representations of three dimensions than people who live in a natural setting. More art, more television, more printed images, and more charts.

This is really bad. Habits are accessible to conscious formation, introspection, and variation. Lower-level heuristics in the visual system, such as those giving rise to the classic "optical illusions", are not. They are very different things, and conflating them is evidence of breathtaking ignorance of his subject on Duhigg’s part.

That was my thought too. The upper line looks longer for the same reason that the tracks look like they're receding into the distance: because if a three-dimensional setting ever produced that particular sight picture, then both of those statements would be true.

Agreed. Habits can be unlearned. Is it possible to unlearn a optical / auditory illusion?

The McGurk Effect works on people even when they know about it. A researcher who has been studying it for more than 20 years still succumbs to it.

I've seen these railroad tracks many times, and my first impression is always that the lines are different sizes. Only when I mentally eliminate the railroad tracks do they lines appear the same length.

The brain is wired to complete a mental picture in the absence of information. Illusionists prey upon this predilection as well as understanding the psychology of misdirection.

Some of these illusions and examples of cognitive bias are invalid precisely because they are contrived. Real world situations are rarely perfectly suited for deception. Our first impressions are usually correct impressions, and wrong impressions don't usually last very long.

On the other hand, eyewitness testimony is notoriously faulty. People often remember seeing what they want to see, what they expect to see, or what they think you want them to see. This is an information processing problem, not a vision or perception problem. An experienced litigator who suspects a cognitive bias can usually raise sufficient doubt in a witness' mind through delicate questioning. A litigator must also know the other factors which degrade channel richness such as noise, light conditions, drugs or alcohol, state of mind, alertness, etc.

I'm quite certain the author suggests that those "lower-lever" heuristics are of the same group of heuristics our brains use to calculate desire, risk, and reward. It is not that the author misunderstands optics, rather the author is stating that these lower-level heuristics play a major role in our actions as well as our visual interpretations.

Agreed. These are mitigated by very different brain structures. Basal ganglia vs. Visual cortex. So distracting I can't even see his main point here.

>a relatively simple task such as hitting a golf ball

Never played golf, have you?

Hitting it is relatively simple. Hitting it and making it go where you want is incredibly complex.

I'm with this Jim. Certainly golf clubs are masterly designed, but 10-year old me was completely infuriated that they wanted me to tee off with a wobbly stick.

@The Other Jim: Hitting a golf ball IS a relatively simple task... it's hitting it well that's the kicker.

@J Storrs Hall: The illusion of three-dimensionality produced in a drawing is a habit. The brain is assuming - and then assembling an internal representation - of a three dimensional scene, without access to the typical cue of a three dimensional scene - parallax. It's a habit of thought, to be sure, but still a habit.

I'd be interested in seeing how people without any experience of three dimensional vision (single eye blindness, etc.) respond to this scene. I once had a girlfriend with this condition, from birth, and I recall that it made many optical illusions odd to her, but that was over a decade ago and I'm unfortunately not in a position to contact her to ask...

Finally, I'm not sure, but I guess I have to call "Reading many of the books recommended by Alex and Tyler" a good habit... at least for Amazon and the authors. Less calories and at least as enjoyable as a cookie, at least!

Read this instead of the Gladwellian popsci...

Zenon W. Pylyshyn (1999). Is Vision Continuous with Cognition? The Case for Cognitive Impenetrability of Visual Perception. Behavioral and Brain Sciences 22 (3):341-365.
Although the study of visual perception has made more progress in the past 40 years than any other area of cognitive science, there remain major disagreements as to how closely vision is tied to general cognition. This paper sets out some of the arguments for both sides (arguments from computer vision, neuroscience, Psychophysics, perceptual learning and other areas of vision science) and defends the position that an important part of visual perception, which may be called early vision or just vision, is prohibited from accessing relevant expectations, knowledge and utilities - in other words it is cognitively impenetrable. That part of vision is complex and articulated and provides a representation of the 3-D surfaces of objects sufficient to serve as an index into memory, with somewhat different outputs being made available to other systems such as those dealing with motor control. The paper also addresses certain conceptual and methodological issues, including the use of signal detection theory and event-related potentials to assess cognitive penetration of vision. A distinction is made among several stages in visual processing. These include, in addition to the inflexible early-vision stage, a pre-perceptual attention allocation stage and a post-perceptual evaluation, memory-accessing, and inference stage which provide several different highly constrained ways in which cognition can affect the outcome of visual perception. The paper discusses arguments that have been presented in both computer vision and psychology showing that vision is "intelligent" and involves elements of problem solving". It is suggested that these cases do not show cognitive penetration, but rather they show that certain natural constraints on interpretation, concerned primarily with optical and geometrical properties of the world, have been compiled into the visual system. ...

If you have more time, read Visual Perception by Bruce, Green and Georgeson. I draw your attention in particular to page 412, the section on modularity and cognitive penetrability.

Paraphrase, please!

A wall of text is extremely unappealing.

Thanks for this ref. I read all of it and have found the source paper. I'd rather be a boll weevil than a snowflake

I've forgotten how to ride a bike.

Not hard to reacquire the habit, though.

Exactly, I rode a bike for years as a kid, often for many miles, I could do tricks too, but then I stopped and ten years later I bought a bike and realized I couldn't even go ten feet, it took me longer to relearn than it took me to learn in the first place.

As a kid, I rode my bike in Toronto and New York City without a helmet, in traffic, against lights and signs, speeding down steep hllls, and over bumps without fear.

Now, riding my bike in a Chicago suburb even with a helmet on produces enormous fear. I obey traffic laws. As a kid, I never worried about car doors opening.

Some days I wonder how I survived youth. I'm worried I'm teaching my kids too much of my fears. I'm wondering if our parents' generation were less frightened, more ignorant, or less loving. :)

Parents today seem hyperprotective.

The name I adopted for this board was "Tyler Fan" but I love it when Alex writes posts like this. Your "Principles" textbook must be pretty kickass; makes me regret my semesters with Mankiw's "Principles."

"Once you learn, however, you never forget": I've found this to be true of cycling but not of soldering.

"Habits can never truly be broken but if one can recognize the cues and substitute different rewards to produce new routines, bad habits can be replaced with other, hopefully better habits."

Like the habit of constantly needing to have a drink in your hand. Replace the booze with tea.

Do the same with coffee. I started this habit long before I was consuming alcohol. Took me many years to understand that it was a social cue I used to let someone else know I'm done talking... Forcing myself to not drink with a cup in hand is insanely difficult and usually prevents me from listening.

>Hitting a golf ball IS a relatively simple task… it’s hitting it well that’s the kicker.

Well then, I suppose the world is utterly filled with simple tasks.

Speaking Latin. Cliff diving. Leg amputation. Anyone can do these things. Doing them well is the kicker.

E Pluribus Unum!

First word of 2nd paragraph. Apostrophe. I read no further.


Wait a second. You posted to tell us how little you read, and then you thanked yourself for your post about two hours later?!

I assume he thanked Alex for fixing the typo.

MR folks should read the classic -- Inner Game of Tennis, by Tim Gallway. This was written back in the 1970's but has become a classic outside of just the tennis realm. He calls the two forces self 1 and self 2. He writes so clearly on performance and its worth the read. its short and fast, only a 100 pages or so, but its well worth it.

Kahneman actually describes system two as lazy and easily tired. Thus it may not be habit that makes you eat the cookie, but rather an inability for system two, due to stress, exhaustion, distraction or whatever, to preempt the system one response.

So will power depletion affects the two systems differently? I can believe it.

And as to the optical illusion, I saw it an immediately knew that answer was that they are the same length. Familiarity with questions and figures of that type generates a sense of recognition and cognitive ease that either overcomes or becomes part of system one (I can't recall how Kahneman phrased it). So is the habit (which I take to mean learned behavior) seeing that top line as longer or knowing that they are both the same?

As someone mentioned above, it would be interesting to see the results for people who are unfamiliar with this type of visual representation to figure out whether the optical illusion is learned or not.

I suspect you *knew* it was a trick. If tested with many such images some of which had equi-length lines and others not could you tell them apart? Does the presence of the confounding tracks interfere so much with visual discrimination to not be able to compensate for it?

re 'habits'.
from the NYTImes Sunday magazine:

February 16, 2012
How Companies Learn Your Secrets

[text excised]

A link would have been great.

Way back it was assumed that 'habits' needed regular reinforcment or you became 'rusty'.
Is not training almost eternal in skills such as sports , singing professionally, ballet etc.

I'm ordering the book.

Comments for this post are closed