Thinking, Fast and Slow

This is an invaluable book that every person who considers him/herself educated should read - even study. Indeed, it is a scandal that mastering the material in this book isn't considered an essential component of a high school education.The author was awarded the Nobel in Economics for his work on what he calls decision theory, or the study of the actual workings of the typical human mind in the evaluation of choices, and the book itself presents the findings of many decades of psychological studies that expose the endemic fallacious thinking that we are all prone to, more or less. The lives of all of us could be improved by lessons learned from this book, not just individually, through self-education, but also on the large scale, if the large scale decision makers in this society in and out of government could be educated as well. In fact, it is largely because these large scale decision makers are no better than the rest of us in their ability to think straight and plan well, that society is as screwed up as it is, and that essentially all of its institutions are diseased and corrupt. The lesson there, however, is that decision making needs to be returned to the individual - that the powers that be need to be deprived of their powers to mess up the lives of the rest of us.Despite the many virtues of this book - it is well-written, engaging, and its academic author reasonably restrained in the tendencies of his tribe to blathering in abstractions - it is a bit disappointing at the very end, when the author proves unable to synthesize all his material into a comprehensive theory of the thinking, and deciding mind - or at least into a set of carefully formulated principles that provide a succinct summary of the principles of human thinking, both typical and ideal.Kahneman uses throughout a construct that implies that we are of two minds: System 1 is the fast-thinking, intuitive, mind, prone to jumping to conclusions; while System 2 is the slow-thinking analytical mind, that is brought into play, if at all, only to critique and validate the conclusions that we have jumped to. System 2, we are told, is lazy, and if often just rubber-stamps the snap judgements of System 1, or if pressed, rationalize them, instead of digging critically as well as constructively into the complex underpinnings of the material and sorting them out as best it can.Instead of working this construct up into a comprehensive model of mind, K merely uses it as a loose schema for representing the kinds of thinking thought to underlie the results derived from the many psychological experiments that he here reports on. This neglect raises the question at many points as to just how well the experimenters have really understood the thinking that underlies the behavior of their subjects. But this, I am sorry to say, is a weakness of virtually all psychological experimentation, which is still just beginning to come to grips with the complexity and varieties of cognitive style of the human mind.What Kahneman does do, however, is to provide convenient labels for many characteristic types of fallacious thinking, although again, the exact role of System 1 and System 2, and their interaction, is inadequately explicated. Instead, towards the end of the book, another, somehow related, but nominally independent theme is developed: the disturbing divergence between the experiencing self and the remembering self. This is in itself such an interesting and important idea, so pregnant with both psychological and philosophical implications, that it could have used a fuller treatment, and again, there is no coherent integration of this theme with the System1/System2 construct.The idea here is that our present experience includes our most salient memories of previous experiences - for example the highlights of past vacations, or out of the ordinary episodes of our lives. Somewhat surprisingly, though, what we remember is a systematic distortion of the actual experience. Our memory collapse the duration of various aspects of our experience and highlights only the peak moment(s) and the final moments, perhaps with a nod toward the initial presentation of the experience. And this systematic distortion of the actual experience in all its fullness, can lead us to make irrational and detrimental choices in deciding whether to repeat the experience in the future. Thus, a bad ending to an otherwise wonderful experience can spoil the whole thing for us in memory, and cause us to avoid similar experiences in the future, even though by simply anticipating and improving the ending we might make the whole experience as wonderful as most of the original was. Likewise, subjects in experiments involving either long durations of pain, or much shorter episodes of pain with a higher peak, were consistently more averse to the latter rather than the former - or they overemphasized the way these presentations ended as a factor in judging them as a whole.These are important findings that go the heart of the question of how best to steer our course through life, but here is the only attempt at integration of this remembering vs experiencing self theme, and the System 1/ System 2 theme that I find in the final chapter, Conclusions:"The remembering self is a construction of System 2. However, the distinctive features of the way it evaluates episodes and lives are characteristics of our memory. Duration neglect and the peak-end rule originate in System 1 and do not necessarily conform to the values of System 2".There is more here, but it merely repeats the earlier analysis of the relevant experiments.No evidence is presented as to the respective roles of System 1 and System 2 with respect to the laying down of memory, to its decay, or with respect to a recently discovered phenomenon: memory reconsolidation. Nor is any account taken of what has been learned, much of it in recent decades, about the interactions between short, intermediate, and long term memory, or any of the radically different modalities of episodic (picture strip) and semantic (organized, abstracted) memory. Consequently, Kahneman's vague reference to the "characteristics of our memory" is essentially a ducking of the question of what the remembering self is. I think that at best, the finding of the replacement of the original experience by an abstract predicated on peak-end bias is an exaggeration, though there's no question that "duration neglect" is in operation, and a good thing too, unless K means by "duration neglect" not just the stretches of minimally changing experience (which have little memorial significance anyway, but even the consciousness of how long the edited out parts were (this distinction was never made in Chapter 35, where the theme of the remembering vs. the experiencing self is first taken up).Speaking for myself anyway, I have a much fuller memory of my most important experiences than Kahneman seems to indicate. Naturally the highlights are featured, but what I tend to remember are representative moments that I took conscious note of at the time, as though making a psychological photograph. I remember these moments also because I bring them up from time to time when I'm thinking about that experience. For example, I'm thinking now of a long distance race I did in 2014 (a very tough half-marathon, with almost 2000' of climbing). I remember: the beginning section as well as the ending section; each of the rest stations; certain moments of each of the major hill climbs; at least one moment from each of the descents; and a number of other happenings during the almost three hour event. For me in this race, the peak experience occurred right at the end, when I all but collapsed, yet managed to stagger to the finish line. That ending does naturally come first to mind as a representation of the entire event, but it is merely the culmination of a long and memorable experience with many moving parts, and if I want, my remembering self can still conjure up many other moments, as well as a clear sense of the duration of each of the sections of the course.Over many years most memories fade, and it's certainly reasonable to suppose that in extreme cases, where they are all but forgotten, only a single representative moment might be retained. However, if we can say anything for sure about memory it is this: we remember what we continue to think of and to use, and we do that precisely because this material has continuing importance to us. The recent research in memory reconsolidation tells us that when we do bring up memories only occasionally, we reinforce them, but we also edit and modify them to reflect our current perspectives, and sometimes we conflate them with other seemingly related knowledge that we've accrued. We are thus prone to distort our own original memories over time, in some cases significantly, but we may still retain much more of the original experience that just the peak and the end, and if we do reinterpret our memories in the light of more recent experience, that's not necessarily a bad thing. In any case, the memories that occur in the present may be said to be a joint project of the experiencing as well as the remembering self, which rather erodes the whole Two Selves concept that Kahneman first posited.I do not mean to criticize the valuable evidential material in the book, and in general I think that Kahneman, and the other researchers and thinkers whom he quotes, have drawn reasonable conclusions from the experiments they report on. But ultimately, the book, as well as the fields both of psychology and brain neurophysiology suffer both in coherence and meaningfulness because they aren't predicated on a more comprehensive theory of mind. It's the old story in science, first formulated by Karl Popper in his 1935 book, The Logic of Scientific Discovery: unless we approach the data with an hypothesis in mind - unless, indeed, we seek out data likely to be relevant to a particular hypothesis, we're not going to make any enduring progress in understanding that data in a comprehensively meaningful way, let alone be able to make falsifiable deductions about elements of the system for which we have at present no data. Popper's quotation from the German philosopher Novalis comes to mind - "Theories are nets: only he who casts will catch."In the final, "Conclusions", chapter, K caricatures the abstract economists' model of homo econimicus (man as a rational optimizer of his utility), contrasting it with the more sophisticated and experientially grounded model of psychologists such as himself. In keeping with his penchant for framing (or spinning) his presentation favorably to his own perspective, he calls the economists' model "Econ", and his own "Human". In fact, "Econ" was never meant to represent man in all his humanity, and Kahneman's Economics Nobel, recognizing his decision theory contributions to economics, was preceded by many other Nobels to economists who had been expanding the concept of the economic actor into psychological territory for decades. In fact, the essential view of the Austrian economists dating from the 1920s (von Mises, Hayek, and their predecessors) is that economics is in the end wholly dependent on psychology because it is predicated on the unknowable, unquantifiable subjective value preferences of humans, acting individually and in concert. Cautious generalizations can perhaps be made about human psychology in general, but I think that on the whole the Austrians have been a bit wiser in their restraint than Kahneman and his many, and mostly lesser, pop psychology compatriots have proved in their often sensationalist extrapolations from lab experiments.Here is an example, I think of Kahneman over-reaching. He speaks repeatedly of the laziness of System 2, and its foot dragging reluctance to get involved in the thinking process, but in the real world, snap judgements are good enough for immediate purposes, and the better part of rationality may be to go with one's fast thinking intuitive System 1: indeed, Kahneman acknowledges this himself in passing, both in his beginning and his ending, but this isn't enough to counterbalance the overall argument of his book.Kahneman also, in his final chapter, speculatively extends his findings into the political sphere (his liberal Democratic Party bias has already been made clear by gratuitous and somewhat annoying usage of salient modern politicians in examples), but not to any great effect.Kahneman advocates "libertarian paternalism" consisting of government programs that people are enrolled in automatically unless they opt out by checking a box on forms - thus manipulating the presentation frame so as to trick them into signing on to what some government bureaucrat thinks is good for them. Of course, as long as people are allowed to opt out, one can't call the choice here anything but libertarian, though to be consistent with their socialist mores, liberals like Kahneman really ought to object to such practices as being manipulative advertising. This libertarian finds nothing objectionable about the way such a choice might be presented - after all, the average man, if adequately educated and prepared for the real world, should have no trouble seeing through the frames. What is not only paternalistic, but totalitarian in spirit, is the extortion of taxpayer money to finance such government programs in the first place.Somehow, it fails to occur to Kahneman that most people could be trained to recognize and avoid fallacious thinking during all those years of enforced and mostly wasteful schooling - just as most people can be trained to recognize the Müller-Lyer illusion for what it is. IMO every high school graduate should be required to learn to recognize and avoid the paradigm cases of fallacious thinking presented in Thinking, Fast, and Slow, and this material could profitably be expanded to cover the many rhetorical tricks used by the manipulators and spinmeisters, both public and private, who batten off of our society. With such training in critical thinking, and with the reintroduction of enough honest and rigor to begin high school graduates up to the 12th grade reading and writing proficiencies that were routine in the 1950s, the need for college as life preparation would be altogether obviated, and most young people could avoid wasting their early years in college, piling up debt, and get on with their work and/or their self-education, as they chose.

Back in 1994, Massimo Piattelli-Palmarini, Director of the Institute of San Raffaele in Milan, Italy, wrote a charming little book about common cognitive distortions called Inevitable Illusions. It is probably the very first comprehensive summary of behavioral economics intended for general audience. In it, he predicted that the two psychologists behind behavioral economics - Amos Tversky and Daniel Kahneman - would win the Nobel prize. I didn't disagree with the sentiment, but wondered how in the world were they going to get it since these two were psychologists and there is no Nobel prize in psychology. I didn't think there was much chance of them winning the Nobel Prize in economics. I was wrong and Piattelli-Palmarini was right. Kahneman won the Nobel prize in Economic Sciences. (Tversky unfortunately prematurely passed away by this time.) Just as Steve Jobs who was not in the music industry revolutionized it, the non-economists Kahneman and Tversky have revolutionized economic thinking. I have known Kahneman's work for quite some time and was quite excited to see that he was coming out with a non-technical version of his research. My expectations for the book were high and I wasn't disappointed.Since other reviewers have given an excellent summary of the book, I will be brief in my summary but review the book more broadly.The basis thesis of the book is simple. In judging the world around us, we use two mental systems: Fast and Slow. The Fast system (System 1) is mostly unconscious and makes snap judgments based on our past experiences and emotions. When we use this system we are as likely to be wrong as right. The Slow system (System 2) is rational, conscious and slow. They work together to provide us a view of the world around us.So what's the problem? They are incompatible, that's what.System 1 is fast, but easily swayed by emotions and can be as easily be wrong as be right. You buy more cans of soup when the display says "Limit 12 per customer". We are on autopilot with this system. System 1 controls an amazing array of behavior. System 2 is conscious, rational and careful but painfully slow. It's distracted and hard to engage. These two systems together provide a backdrop for our cognitive biases and achievements.This very well written book will enlighten and entertain the reader, especially if the reader is not exposed to the full range of research relating to behavioral economics.This book serves an antidote to Malcolm Gladwell's Blink. Although Gladwell never says that snap judgments are infallible and cannot badly mislead us, many readers got a different message. As the Royal Statistical Society's Significance magazine put it "Although Gladwell's chronicle of cognition shows how quick thinking can lead us both astray and aright, for many readers Blink has become a hymn to the hunch." While Kahneman does show how "fast thinking" can lead to sound judgments, he also notes how they can lead us astray. This point is made much more clearly and deliberately in Kahneman's bookAll my admiration for the brilliance and creativity of Kahneman (and Tversky) does not mean that I accept 100% of their thesis. Consider this oft-quoted study. Linda is 31 years old, single, outspoken, and very bright. As a student, she was deeply concerned with the issues of discrimination and social justice, and she also participated in anti-nuclear demonstrations. Which is more probable?1. Linda is a bank teller.2. Linda is a bank teller and is active in the feminist movement.Eighty-five percent of test subjects chose the second option, that Linda was a bank teller and active in the feminist movement. Kahneman's interpretation is that this opinion is wrong because the probability of a (random) woman being a bank teller is greater that than person's being a bank teller AND a feminist. What Kahneman overlooks here is that what most people answered may not be the question that was asked. The respondents may not have been concerned with mathematical probabilities, but rather could be responding to the question in reverse: Is it more likely for a current activist to have been an activist in the past compared to others in the profession? A more formal and theoretically better argued rebuttal of some of Kahneman's hypotheses can be found in the works of Gerd Gigerenzer.Kahneman notes that even top performers in business and sports tend to revert to the mean in the long run. As a result, he attributes success largely to luck. I'm not so convinced of this. There can be alternative explanations. People who achieve high degree of success are also exposed to a high degree of failure and the reversion to the mean may be attributable to this possible mirror effect. Spectacular success may go with spectacular failure and run-of-the-mill success may go with run-of-the-mill failure. Eventually everyone may revert the mean, but the ride can be very different. Chance may not account for that.Another concern is that much of the work is done in artificial settings (read college students). While much of what we learnt can perhaps be extended to the real world, it is doubtful every generalization will work in practice. Some may find Kahneman's endorsement of "libertarian paternelism," not acceptable. More importantly, when applied to the real world it did not always found to work.In spite to these comments this book is written carefully in a rather humble tone. I also appreciated Kahneman's generous and unreserved acknowledgement of Tversky's contributions and his conviction that, had he been alive, Tversky would have been the co-recipient of the Nobel Prize. My cautionary comments probably have more to do with the distortions that might arise by those who uncritically generalize the findings to contexts for which they may not applicable. As mentioned earlier, the wide misinterpretation of Gladwell's Blink comes to mind.Nevertheless, Thinking Fast and Slow is a very valuable book by one of the most creative minds in psychology. Highly recommended. For a more complete and critical understanding, I also recommend the writings of the critics of behavioral economic models such as Gerd Gigerenzer.PS. After I published this review, I noticed an odd coincidence between Thinking Fast and Slow and Inevitable Illusions that I mentioned in my opening paragraph. Both books have white covers, with an image of a sharpened yellow pencil with an eraser top. How odd is that?