The Greatest Trap of Overconfidence — Moments of Losing the Self

The previous series, "Cultivating Confidence" in ten volumes, drew the map of assembling confidence in three dimensions. This series begins with the paired proposition — for one who has gained confidence, the gravest threat is not external difficulty but one's own "overconfidence". Daniel Kahneman, Don Moore and Paul Healy's typology of overconfidence (2008), Philip Tetlock's long-term study of expert judgement, the self-knowledge wisdom of Laozi and Confucius, the observation Darwin left, and the history of pharmaceutical disasters — seven perspectives have pointed, in different words, to the same place. Overconfidence is deeper than ignorance, broader than the lack of capacity, and the more easily entered the more one is an expert. This series is especially important for the reviewer — taking ten volumes to dig into the structural risk that one believes excessively in the correctness of one's own judgement and overlooks the opportunity for correction. This volume opens the series by depicting why overconfidence is called "the greatest trap."

01Why Overconfidence Is Called "the Greatest Trap"

Overconfidence is the phenomenon that psychology, economics, and management research, since the modern era, have all consistently placed as the most lethal bias. Why is it called the greatest trap? There are three reasons.

This series is indispensable for the reviewer precisely because of the third reason — judgement on regulatory conformity ripples into the decisions of the requester, the originating client, and ultimately the patient and the clinician. To believe excessively in one's own judgement there is the root structure of the failures the history of pharmaceutical harm has taught repeatedly. After honing confidence in the previous series, the training to recognise overconfidence is precisely what is needed.

02The Definition of Overconfidence — Moore & Healy's Three Types

The one who defined overconfidence academically and precisely is the Carnegie Mellon University psychologist Don Moore (1965–) with Paul Healy. In their 2008 paper in Psychological Review, "The trouble with overconfidence," they classified overconfidence into three types. This is the standard framework of contemporary overconfidence research.

"Overconfidence is not a single phenomenon; there are at least three different kinds. (1) Overestimation — estimating one's capacity, performance, or control higher than it actually is. (2) Overprecision — placing excessive trust in the accuracy of one's judgement (the confidence interval is too narrow). (3) Overplacement — believing oneself superior to others (the better-than-average effect). The three are independent, arising by different mechanisms." — paraphrase of Moore & Healy, Psychological Review (2008)

This three-type distinction is decisively important for the practical observation of overconfidence. In the reviewer's field too, the three appear in different forms.

This series will consciously distinguish these three forms of overconfidence as it proceeds. Each demands a different correction.

03Kahneman's Warning — Overconfidence Is the Most Lethal Bias

The Nobel laureate Daniel Kahneman (1934–2024), in 2011, brought together half a century of research on human judgement in Thinking, Fast and Slow. The bias he warned of most in that book is overconfidence.

"Among the biases of human judgement, there is none so persistent, so resistant to correction, and so serious in its consequences as overconfidence. Both experts and non-experts are not immune from overconfidence. And, astonishingly, learning about overconfidence does almost nothing to reduce overconfidence itself. This is what makes overconfidence the most lethal bias." — paraphrase of Daniel Kahneman, Thinking, Fast and Slow (2011), Chapter 24

Kahneman's warning shows directly the motive of this series. To know about overconfidence and to reduce overconfidence are different things. For just this reason, this series aims not at "the transmission of knowledge" but at the habits of cognition that can be embedded in daily practice. The practices for reducing overconfidence treated in Vols. 8 and 10 are the most practical core of this series.

04Expert Overconfidence — Tetlock's Long-Term Study

The political psychologist at the University of Pennsylvania, Philip Tetlock (1954–), measured the accuracy of experts' predictions in a twenty-year study without precedent. From 1984 to 2003 he gathered some 28,000 predictions from 284 experts and compared their accuracy with actual outcomes.

"The conclusion of twenty years of research was clear. Experts' predictions were not much different from random guessing in most fields. Yet experts expressed high confidence in their predictions. The correlation between confidence and accuracy was weak — often negative. The more confident the expert, the more likely they were to be wrong." — paraphrase of Philip Tetlock, Expert Political Judgment (2005)

Tetlock's discovery was shocking. Being an expert itself produces overconfidence. Experience produces confidence; confidence produces certainty in the prediction; certainty takes away the opportunity for verification — this loop distorts the expert's judgement. The implication for the reviewer is deep — the longer one's years of experience, the greater the risk of overestimating one's own judgement. This is not a personal problem but a structural feature of human cognition.

Tetlock, in his 2015 Superforecasting, analysed the characteristics of the rare people — superforecasters — who can reduce the influence of overconfidence. What they have in common is not the quantity of knowledge but the habits of thinking. Always considering several perspectives; not fearing the updating of belief; writing out candidly the basis of their predictions. These connect directly to the practices treated in the final volume of this series.

05Laozi and Confucius — The Eastern Wisdom of Self-Knowledge

The importance of recognising overconfidence was not discovered by contemporary psychology. The ancient Chinese sages were pointing to the same place 2,500 years ago. Laozi (c. 6th c. BCE), in Chapter 33 of the Tao Te Ching, says:

"He who knows others is wise; he who knows himself is enlightened. He who conquers others has force; he who conquers himself is strong." — Laozi, Tao Te Ching, Chapter 33

What Laozi shows is the proposition that to know others is wisdom (zhi); to know oneself is enlightenment (ming). Wisdom and enlightenment are not the same. Wisdom is the capacity to analyse an object; enlightenment is the capacity to see through one's own bias. The implication for the reviewer — to know the regulation, the industry, the requester is "wisdom." To know one's own habits of judgement is "enlightenment." They are different capacities; the latter is the harder to acquire.

His contemporary Confucius (551–479 BCE) pointed to the same place. The Analects, Book II "Wei Zheng," has the famous proposition:

"To say of what you know that you know it, and of what you do not know that you do not know — this is knowledge." — Confucius, Analects, Book II "Wei Zheng"

Confucius's proposition is short and deep. To say "I know" of what one knows and "I do not know" of what one does not know — this is genuine "knowing." He was already pointing to the very middle of the trap of overconfidence. Not being able to distinguish what one knows from what one does not know — this is the heart of overconfidence, and Confucius's "to know not-knowing as not-knowing" is the most concise prescription against it.

06Darwin's Observation — Ignorance Generates Confidence

The nineteenth-century British naturalist Charles Darwin (1809–1882), who revolutionised natural science, also left sharp observations on human psychology. In the introduction to The Descent of Man (1871), a passage often quoted runs:

"Ignorance more frequently begets confidence than does knowledge: it is those who know little, and not those who know much, who so positively assert that this or that problem will never be solved by science." — Charles Darwin, The Descent of Man (1871), Introduction

Darwin's observation is the precursor of the contemporary Dunning-Kruger effect (treated in detail in Vol. 2). As knowledge grows, one comes to notice the breadth of what one does not know. Conversely, the person with little knowledge does not see the breadth of what they do not know — "they hold a map with the four corners hidden." That one cannot see the boundary of one's own knowledge is the structure that produces overconfidence.

The implication for the reviewer is direct. When one enters a new domain — an unfamiliar therapeutic area, a new regulation, a novel medium — what is most dangerous there is the "I have roughly understood" feeling at the early stage. How much unknown remains in the depths is not yet visible. Genuine understanding lies beyond the recognition of the breadth of what one does not know.

07The History of Pharmaceutical Industry Overconfidence — What the Disasters Teach

As a special context of this series, let us briefly survey the history of overconfidence in the pharmaceutical industry. Several of the major disasters of the twentieth century — on the side of manufacturers, regulators, or clinicians — had overconfidence playing a decisive role.

These cases have their individual causes, but they share a structure. "The confidence interval of judgement looked narrower than it actually was." Of Moore & Healy's three types, Overprecision in particular has run through the history of pharmaceutical harm. The ethical foundation of this series for the reviewer lies here — reducing overconfidence is connected directly to the central industry ethic of patient safety.

08Overconfidence Goes Beyond the Individual — Wrecks Organisation and Relation

To treat overconfidence as a problem of individual psychology is only half right. Overconfidence is a phenomenon amplified within organisation and relationship. The groupthink set out in 1972 by the American social psychologist Irving Janis (1918–1990) in Groupthink systematised the organisational amplification of overconfidence.

"A cohesive group, from the pressure to preserve consensus, suppresses critical judgement. It becomes hard to express minority views; external information is underestimated; and the whole group falls into the sense 'we are right.' This is groupthink, the source of fatal judgement errors." — paraphrase of Irving Janis, Groupthink (1972)

The implication is clear. The practice of reducing individual overconfidence and the practice of reducing organisational overconfidence are paired. This series primarily treats the practice of individual cognition, but Vol. 6, "Why one cannot learn from failure," and Vol. 7, "Between overconfidence and humility," will bring the organisational context within view. Because the reviewer judges not alone but within a relationship with team, supervisor, and requester, resistance to organisational overconfidence is required.

09In the Pharmaceutical Workplace — What Overconfidence Means for the Reviewer

Let us bring the abstract down to the field. Concrete scenes in the reviewer's daily work in which overconfidence is tested.

10The Map of This Series — Composition and Reading of the Ten Volumes

Finally, let us make explicit the structure of the ten volumes of this series. As with the previous series, ten volumes decompose overconfidence in three dimensions.

Three cautions for reading the series.

First, overconfidence does not entirely disappear. As Kahneman himself acknowledged, overconfidence is a structural feature of human cognition, not extinguishable by knowledge. The aim of this series is not "erasure" but the acquisition of the habit of "monitoring and correcting."

Second, reducing overconfidence is not losing confidence. The previous series, "Cultivating Confidence," and this series are not in opposition. Rather, genuine confidence lasts only when it coexists with self-monitoring against overconfidence. The two series are paired practices supporting both sides of Aristotle's mean.

Third, handling overconfidence is not self-negation. The training to recognise overconfidence is not continually doubting oneself, but possessing the confidence interval of judgement accurately. Not "I do not understand," but recognising with precision "this I can see up to here, but from there onward I cannot."

Overconfidence is deeper than ignorance, broader than the lack of capacity, and the more easily entered the more one is an expert — this is the starting point of the series. Moore & Healy's three types (overestimation, overprecision, overplacement), Kahneman's warning, Tetlock's expert study, Laozi's "he who knows himself is enlightened," Confucius's "to know not-knowing as not-knowing," Darwin's observation, the history of pharmaceutical disasters — the seven perspectives have pointed, in different words, to the same place. Overconfidence does not entirely disappear. For this very reason, the training to recognise it is needed. This is the central proposition of the series.

The bodily techniques, cognitive levers, and negotiation practices of the previous series, "Cultivating Confidence," work as a pair with this series. Confidence and overconfidence are continuous. The one who can handle both at once becomes the genuine professional. The next volume (Vol. 2) treats the most famous experimental ground of overconfidence — the Dunning-Kruger effect and the blind spot of metacognition it reveals.