Don’t Believe Everything You Think

UnknownDaniel Kahneman’s book, Thinking Fast and Slow says there are two distinct systems operating inside your skull. “System 1” gives quick, intuitive answers to questions confronting us, utilizing thought algorithms rooted deeply in our evolutionary past. “System 2” is slower and more analytical, used when we actually (have to) think, as opposed to just reacting.

imagesBecause you utilize System 2 consciously, whereas System 1 works unconsciously, you tend to see yourself embodied in System 2. We do like to believe we do our own thinking, rather than having some black box, to which we have no access, just handing us answers. But the latter is closer to the truth, most of the time. In fact, as Kahneman stresses, System 2 is lazy, hence often glad to just accept System 1’s answers, not even realizing it.

This comports with Jonathan Haidt’s metaphor in The Righteous Mind: System 2 is a rider on the back of an elephant that is System 1. images-1We imagine the rider is in charge. But mostly the rider is really working for the elephant, rationalizing the elephant’s choices.

All this would be fine if System 1 were totally rational, but of course it’s not, and books like Kahneman’s have been taken as debunking the very idea of our being rational creatures. Unknown-1Kahneman uses the term “Econs” (perhaps short for homo economicus) for the hypothetical people who behave as economic theory says. As opposed to – well – “humans,” who do not.

One key example of how System 1’s decisional algorithms are irrationally biased is undue loss aversion, weighting potential losses more heavily than equal potential gains. Unknown-2If offered a coin flip bet paying $10 for heads but costing $6 for tails, most people will refuse; the 50% chance of winning $10 is not enough to compensate for the fear of the pain of losing $6! Lab experiments consistently confirm many permutations of this irrational bias.

We don’t often encounter coin flip bets, but this bias infects many aspects of human behavior – like investment decisions – as well as public policy. Case in point: GM food. Europeans in particular are so averse to potential risks (an extreme “precautionary principle”) that the truly small (indeed, mostly imaginary) risks of GM foods blind them to the truly large benefits.

Meanwhile, the idea that humans aren’t rational has entered political debate, as an argument against market economics, which supposedly is premised on rational economic behavior (homo economicus again).

Here’s what I (System 2) think. Obviously, we don’t behave with perfect rationality. But rationality isn’t either/or, it’s a spectrum, and on the continuum between perfect rationality and perfect irrationality, we’re far toward the rational end. Our entire civilization, with all its complex institutions and arrangements, is a supreme monument to rationality. Unknown-3And as individuals we behave rationally most of the time – overwhelmingly. If you want toast, you put bread in the toaster. That’s rational – as distinguished from, say, praying to a toast god. (And we’re getting ever better about this.) Furthermore, your preference for toast over cereal is a rational choice, based on your long experience of what is most likely to please you. You even know how toasted you like it.

And even when we default to System 1, that is not irrational. Let’s not forget that System 1 evolved over many eons not to lead us astray but, instead, to help us cope with life’s challenges (thus to survive and reproduce; for instance, a loss aversion bias made a lot of sense in an environment where “loss” could well translate as death). So – for all its biases and quirks, extensively explicated by Kahneman – System 1 also has a lot of virtues. In fact we simply could not function without it. If we had only System 2, forcing us to stop and consciously analyze every little thing in daily life, we’d be paralyzed. Thus, utilizing our System 1 – faults and all – is highly rational.

The same answer refutes the critique of market economics. We are far more rational than not, in our marketplace choices and decisions concerning goods and services. Market actors are fundamentally engaged in serving their desires, needs, and preferences, in as rational a manner as could reasonably be expected, even if imperfect. (See my review of Tim Harford’s The Undercover Economist.) Allowing that to play out, as much as possible, is more likely to serve people’s true interests than overriding their choices in favor of some different (perforce more arbitrary) process.

Kahneman was informative about a topic of perennial interest – how people form and maintain beliefs.* Here again, while we fancy this is a System 2 function, System 1 is really calling the shots; and again is reactive rather than analytical. System 1 jumps to conclusions based on whatever limited information it has. Kahneman uses a clumsy acronym, WYSIATI – System 1 works as if “what you see is all there is” – i.e., there’s no additional information available – or needed. System 1 is “radically insensitive to both the quality and quantity of the information that gives rise to impressions and intuitions.”

Unknown-4What’s most important to System 1 is that the story it creates be coherent; it’s averse to the discomfort of cognitive dissonance, and hence hostile to any new information that doesn’t jibe with the story it has already created. Indeed, it is paradoxically easier to construct a coherent story the less you know – fewer pieces to fit into the puzzle. Experiments have shown that subjects exposed to only one-sided information – and who know that that’s so – nevertheless show greater confidence in their resulting judgments than do subjects getting both sides. We have a great ability to ignore our ignorance!

At the risk of sounding smug, I have always sort of recognized this and consciously try to avoid it, by adhering to what I call my ideology of reality. That is, I try to let my perceptions of reality dictate my beliefs, rather than letting my beliefs dictate my perceptions of reality. I am not a perfect “econ,” but I think I am one of the more econ-like humans around.

* An aside: humans are pre-programmed for belief. Is that a lion lurking? The believer loses nothing if he’s wrong. The skeptic, if he’s wrong, may be lunch. Thus belief is the preferred stance, and people readily believe in UFOs, homeopathy, and God.


4 Responses to “Don’t Believe Everything You Think”

  1. Andrew Semeiks Says:

    Based on your review I have no interest in reading this book because of the puerility of it. It is absurd to believe that there are two clearly delineated thought systems in each of us without also denying that behavior can be modified by learning, training and repetition. This would be illogical. People train all the time so that it becomes reflexive, an act not requiring thinking. At the simplest level take tying your shoes, you don’t think about it, also driving a car. At the other extreme would be music instrument playing, scholarly learning, and military training where the reflexive action is drilled into the practitioner and autonomic type response is expected.

    Why not accept that the brain is complex and not try to pigeon hole it into these false little slices?

    The author’s argument is akin to the one of free will. To my thinking you had totally free will when you popped the toast into the toaster. You had the option of not doing it and any discussion about such stuff that your life experiences would have predicted that you would do so is just nonsense.

    As to your coin flip example, on one flip only, I would never take the bet but I would gladly take it on each flip in a total of twenty flips. Any lesser number of flips I would have to think about and would depend on how I feel at the time. Why would it be illogical not to bet $6 for a possible gain of $10 on one flip only. To refuse the bet seems imminently logical to me.

  2. rationaloptimist Says:

    Maybe my review gave a false impression of the book, which actually is not so much about proving there are two distinct systems but, rather, about delineating all the various kinds of innate biases normal humans harbor — like the one exemplified by the coin flip. That you would not take the $6/$10 bet with 50-50 odds shows that you in fact suffer from this irrational bias (and it is indeed irrational).

  3. Andrew Semeiks Says:

    This is for one coin flip only, not multiple flips. It is not illogical if I cannot afford to lose the $6 and the extra $10 is not really needed.

    I just flipped a coin three times and it came up tails three times. Continuing to ten flips it came up 7 tails and 3 heads. If I had bet on heads all 10 times I would have lost $42 dollars and won $30. If I stopped at the first three flips I would have lost $18 and if I owed my bookie $6 my kneecaps would now be hurting.

    I continued flipping for a total of 20 flips and it came up 12 tails and 8 heads. If I kept betting heads I would have lost $72 and gained $80 for a net gain of $8. As I said for 20 flips It would take the bet but not necessarily for less.

    Obviously if by luck I bet tails all times I would be way ahead.

    I tried 20 flips with a different coin and it came up 7 tails and 13 heads.

  4. rationaloptimist Says:

    Andy, c’mon, you’re a smart guy, you didn’t need to actually test flipping a coin to know the odds are 50-50. You’re also smart enough to know the $10/$6 bet is a no-brainer. In fact, that bet has a calculable value — I would pay you $3 for the right to make that bet!!
    “Cannot afford to lose the $6” is a special case which I am pretty sure doesn’t apply in your case.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s