Einstein famously said, “A new type of thinking is essential if mankind is to survive and move towards higher levels”. We need to rethink Psychology if we have to give the world better tools for judgment. Because Psychology - the scientific study of mind and behavior - has stopped relating to Science. As hyperbolic as this may sound, a lot of Science and Physics connected to the human mind predates the subject of Behavioural Finance by more than a century.

“A new type of thinking is essential if mankind is to survive and move towards higher levels”.

Einstein

John Rae wrote about mechanisms of psychology back in 1834 [1], but Science had not advanced enough for the world to recognize this brilliance, just like it took 60 years for the world to recognize Louis Bachelier. Whether we can eventually recognize that the world of psychology is scientific, is not a question of plausibility and philosophy, it's a question of accepting that Behavioral Finance should not have hesitated to acknowledge and accept that it has digressed far away from Science. Richard Thaler’s chosen title, “End of Behavioral Finance” [2], indeed seems to be an apt title for a subject that earned three Nobel Prizes [2002, 2013, and 2017], but struggles with the lack of Science.

Daniel Kahneman, Oliver Sibony and Cass R. Sunstein’s book “Noise” makes another attempt to bring meaning to human judgment, but ends up reinforcing human incapabilities, which are constrained by conflicts, operate in the ineffable but sometimes lucid context of information, within diverse groups that amplify - converge and show patterns of persistence - reversion - dependence, and independence. The book spurns statistical solutions and suggests model thinking to be good, but not good enough to displace human judgment. At a causal level Kahneman et al. wants us to believe that flawed human judgment needs guidelines, information reduction, and a conscious manual intervention to improve judgment, but (un)knowingly he details out, yet again, Rae’s mechanism of psychology, while the giant remains in the background, an unsung hero of psychology.

Noise is a despised reality in Kahneman et al.’s world, which he articulated in “Thinking, Fast and Slow” [3] and with “Noise” [4] he reinforces the difficulty of eliminating noise, at some level accepting that Stranger [5] or Demon [6] is not for psychologists to investigate. The search for elusive machines, that can take a system, and fine-tune its mechanisms, to build a machine that can deliver outcomes like better judgments, is science fiction for most psychologists today.

"Noise is a despised reality in Kahneman et al.’s world"

Kahneman et al.’s vision of human judgment is a landmine of impossibilities, challenges, and contradictions. The variability in judgment leads to rampant unfairness and loss of money. How much one may try, a man can never learn the true lesson about the ubiquity of system noise, which is universal to physiology and psychology. The human mind was imperfect and could only survive by avoiding large errors. Even when faced with the same facts on two occasions, it could not produce identical judgments. No tight control setting could unravel the mystery of the noise. If one for a moment thought that noise could be eliminated, there was the intrinsic variability of the functioning of the brain, which threw a spanner and affected the quality of our judgments, in ways, that one could not hope to control. Any little hope one may have, was dashed by the unending complexity of our neurons which never operated in the same way. The things on which the future depends can simply not be known and our objective ignorance affected not only our ability to predict events but even our capacity to understand them. Noise tended to be unpredictable, invisible, and indescribable because causes were natural, predictions that match the evidence were hard to resist and statistics was difficult. And above all this, teaching people to avoid bias was hard because most time, the directions of biases were not known in advance and humans suffered from bias blind spots. Despite all these impossibilities and challenges, human judgment was irreplaceable, a real-time detection of bias was required, to prevent potentiation errors before they occur and the only way ahead was to educate humans about biases.

"How much one may try, a man can never learn the true lesson about the ubiquity of system noise, which is universal to physiology and psychology."

If the contradictions were not enough, the authors goes and adds a list of conflicts to his case of poor human judgment. Human beings needed to stand out and stray away from consensus, otherwise they risked remaining obscure. In nature, selection only worked with a variation. Humans also needed to keep up the facade of an agreement [an illusion] because, without this naive realism, they won’t be social beings after all. While doing all this flitting from not being obscure to living the illusion of an agreement, humans had little opportunity to notice that their agreed rules were vague. The fast system was on one side driving human biases but on the other side taking away their voluntary capability to categorize.

"The fast system was on one side driving human biases but on the other side taking away their voluntary capability to categorize."

Disagreements were everywhere. Even good arguments could point in opposite directions. Fingerprint examiners and physicians not only disagreed with themselves, but [they did so less often than they] disagreed with others. Even if there were ways to de-bias, the initial reasonable assumption sometimes turned out to be wrong. Conflicts were everywhere.

"Conflicts were everywhere."

In the age of information, the devil is sometime in the detail, sometimes it hides in the details, and sometimes it just vanishes. The authors of “Noise” are unclear about the role of information in judgment. Information could be identifiable, could be rich, helpful, and complex, but never perfect. Bad judgment was much easier to identify than good judgment, however, it never helped in the quantification of the disagreement. Some information was relevant, and some was not. More information was not always better and did not lead to better predictions. And new information, unless it was decisive, provided more opportunities for confusion. Then there was the influence of causal thinking on the information. Causal thinking created stories and made information feel entirely explainable, indeed predictable. The human mind could easily turn a correlation into a causal and explanatory force that allowed humans to make sense of the far less predictable world. Then there was the mood, which changed how humans thought. A good mood could make humans receptive to bullshit [7]. Bad mood influenced eyewitnesses to avoid false testimony. The mood could also influence moral judgments as cited by the footbridge problems [8]. Then there was manipulation of information. Sequencing information, and sharing it, one part at a time, could assist in coherence because irrelevant information exposed early in the judicial process could confound the judgment. The asymmetry between favorable and unfavorable predictions disappeared when more information was available. 

"In the age of information, the devil is sometime in the detail, sometimes it hides in the details, and sometimes it just vanishes."

For Kahneman et al., Information in the true sense did not matter, the key driver was the human ability to behave differently at different times. As their mood varied, some features of their cognitive memory varied with it. Weather information did not directly affect decisions but modified the decision maker’s mood and hence the “Clouds made the nerds look good” [9]. 

"For Kahneman et al., Information in the true sense did not matter, the key driver was the human ability to behave differently at different times."

As I plowed through the Nobel laureate’s discourse on the impaired human mind, my mind kept looking for a needle called Science. I was looking for a mechanism. Eventually, I crossed the bridge and reached John Rae’s mechanism, where a human flitted constantly between states of instant and delayed gratification. And as time passed and the pleasure came closer, it became harder to imagine, what will prevail, the pleasure-driven fast impulsive state of instant gratification or the slow, passive, rationalizing, delayed gratification state. Psychologists have not given due credit to John Rae and his work on intertemporal choices, but the noise of his mechanism continues to create unpredictability for the subject. 

"I crossed the bridge and reached John Rae’s mechanism, where a human flitted constantly between states of instant and delayed gratification."

Kahneman et al.’s discussion on structures, mechanisms, groups, boundaries, behaviors, dependence and independence along with noise rekindled a new hope in me, a hope of a scientific resolution regarding the predicament of human judgment, which though erroneous, had limited assistance from Science for clarity and guidance. The authors’ hopelessness regarding the human mind is rooted in his reluctance to appreciate the mechanistic nature of psychology. His probable openness to rearrange the pieces of his puzzle could allow him to understand the beautiful Nature at work. Where instead of elements, facts and judgments, he could see a mind machine operating perpetually. A machine with distinct and polarized parts. A machine with contradictory expressions of wisdom vs. madness, streaks vs. balance, distortion vs. order, bias vs. error, comparison vs. singleness, dependence vs. independence, ex-post vs. ex-ante, etc.

“Noise” details the Delphi method of social aggregation and claims that the learning worked because at some level it could reduce noise and therefore error. However, this could only happen in a state of independence and when the crowd did not share biases. Independence was a prerequisite for the wisdom of crowds. If people were not making their judgments and were relying instead on what other people think, crowds might be not so wise after all. A first move could set the future trend. Group manipulation was easy as popularity and unpopularity were self-reinforcing. Several studies done with the manipulated ranking of songs confirmed the herding behavior. A similar effect was seen in political positions. Groups could amplify noise. They could go in all sorts of directions and could have wildly disparate rankings. Social influence reduced group diversity without diminishing collective error. Multiple independent opinions when properly aggregated could be strikingly accurate but even a little social influence could produce a kind of herding that undermined the wisdom of crowds.

"If people were not making their judgments and were relying instead on what other people think, crowds might be not so wise after all."

The same mind that was producing flawed judgment was also remarkably replete with patterns of persistence over time and reversion [a change of course]. Studies showcased that human beings leaned toward restoring a form of balance. After a streak or a series of decisions that go in the same direction, they were more likely to decide in the opposite direction than would be justified, as a result, errors, and unfairness were inevitable. 

"After a streak or a series of decisions that go in the same direction, they were more likely to decide in the opposite direction."

Though there was a distinct behavior at work between thinking states, psychologists including Kahneman see it as systematic over and underestimation of the probability that certain changes from the statuesque will occur. This is Gambler’s fallacy. On one side is a streak and on the other side there is balance.

"On one side is a streak and on the other side there is balance."

“Noise” boldly described that consistent bias produced a consistent error. From a linear Bayesian facet, yes, it is hard to disagree with this statement. Because, the more the bias, the more the extremity and hence more the error. But Kahneman et al.’s model of the human mind does not have a boundary. Extremity is an error that does not know how to control itself. This is why emotion outrage was believed to be a primary determinant of punitive intent. 

"Extremity is an error that does not know how to control itself."

Unlike the linearity depicted in “Noise”, John Rae’s mechanism flits between two non-linear states. If we extend the logic, the mind cannot live in one state of impulse, or one state of self-control and delayed gratification. The mind always existed in dual, interconnected, self-balancing states. The more the imbalance, the more the innate ability of the mind to seek balance. In such a mechanism, consistent bias did not produce a consistent error, rather it produced the needed imbalance, for self-correction and learning. 

"The mind always existed in dual, interconnected, self-balancing states."

The human mind is a learning mechanism. This is why AI has a hard time figuring out how to make human-like AI. To call the human mind unsound in its capability to judge is like giving up on a smoker, telling an obese person, he will never lose weight, or abandoning an alcoholic. A society is capable to govern itself better, despite its failings.

"The human mind is a learning mechanism."

Kahneman et al. believes that bias and noise are independent, but he does not ask, why do people have a remarkable view to match intensities. Where does the sense of proportion come from? Why despite manipulability, do some distinct orders prevail? How does the instant gratification mind, has an innate ability to conjure up self-control?

"Kahneman et al. believes that bias and noise are independent, but he does not ask, why do people have a remarkable view to match intensities."

Meanwhile in Rae’s world, bias and noises are dependent. Bias nurtures error, which creates extremities in the mind, which self-regulates and balance the error, on occasions swinging to the other extreme, perpetually, while gaining experience, knowledge, and intelligence. 

"In Rae’s world, bias and noises are dependent."

The irony of the message in “Noise” is that though models and mechanistic elements are strewn across the text, they are not seen as a resolution to the faulty judgment. According to the text, guidelines cut the noise and made decisions unacceptably mechanical. Dialectical bootstrapping produced improvements in accuracy. The simple frugal model beat human judgment. Simple mechanical rules were generally superior to human judgment. Trivial formulas, mechanical aggregation, consistently applied, outdid clinical judgments. Models of judges consistently outperformed the judges they modeled. The better models had a defining characteristic that the same rule applied to all cases. Models with simple rules that could be applied with little or no computation had produced impressively accurate predictions in some settings. But even with all this proof the book calls the models marginal enhancements that could never breach algorithmic aversions as humans expected a lot more from machines.

"Trivial formulas, mechanical aggregation, consistently applied, outdid clinical judgments."

The human heart is not placed symmetrically between the human lungs. Two arteries feed the lungs from the left side of the heart and three arteries feed the lungs from the right side of the heart. Why did nature create this asymmetry? Is it a conscious design or a design flaw? There are many anatomical answers to this question. But one could speculate, that asymmetry was designed for instability, a necessity for building dynamic systems. Bias and noise are not independent, they are interconnected, feedback and feed forward into each other. 

"The human heart is not placed symmetrically between the human lungs."

Noise either cancels itself or amplifies bias, pushes it to an extreme, where bias has two choices, either to keep persisting or balance the mechanism by self-correcting itself. The role of noise in this mechanism is not to confound the judgement, but it is to help formulate judgement, as the mechanism processes and computes information. Like the human constraint with memory, the mechanism is also computation light and can process limited information at a certain time. The act of generating extremities and dynamism allows the mechanism to stay lean and light, retaining what’s essential learning. This is why a common frame of reference works so effectively as it simplifies the comprehension of information, despite its causalities, correlations and complexity in judgment.

"Noise either cancels itself or amplifies bias, pushes it to an extreme, where bias has two choices, either to keep persisting or balance the mechanism by self-correcting itself."

Figure 1: Noise and Bias Mechanism

There is not one but many ways to conceptualize nature. And psychologists like Kahneman refuse to step out to their narrow definitions of psychology, of what they believe is non-scientific. Psychology like Science belongs to all of us and as a knowledge society it is our job to question, challenge and seek answers, a new type of thinking. Fortunately for us, there is a lot of Science still to be unearthed from History. The Physics of Psychology explains many of the errors in our judgement and gives a new way for the society to improve its judgment, and not be lost in the linearly defined noise.

"The Physics of Psychology explains many of the errors in our judgement and gives a new way for the society to improve its judgment, and not be lost in the linearly defined noise."

Bibliography

[1] Rae, J. The Sociological Theory of Capital, 1834

[2] Thaler, R. “End of Behavioral Finance”, Financial Analysts Journal, 1999

[3] Kahneman, D. “Thinking Fast and Slow”, FSG, 2011 

[4] Kahneman, D. Sibony, O. Sunstein, C. “Noise”, Little, Brown and Company, 2021

[5] Pal, M. “Mechanisms of Psychology”, SSRN, 2022

[6] Maxwell, J. C. “Maxwell’s Demon”, 1867

[7] Pennycook. G, Cheyne. J. A, Barr. N, Koehler. D. J, Fugelsang. J. A, “Bullshit Receptivity Scale”, 2015

[8] Mikhail, J. “Moral Cognition and Computational Theory”, MIT Press, 2007

[9] Simonsohn, U. “Clouds make nerds look good: field evidence of the impact of incidental factors on decision making”, Behavioural Decision Making, Wiley, 2006