An interviewer can leave a room with a confident impression of a candidate. Three vivid answers seem to fit together; the conversation appears to have revealed a character. Months later, the prediction may prove unreliable. What felt like knowledge was partly the coherence of a story assembled from limited evidence.
Daniel Kahneman spent decades examining the distance between the information a mind possesses and the certainty it feels entitled to claim. With Amos Tversky, he studied the shortcuts of judgment, the weight of losses and gains, and the forms in which a decision becomes imaginable. His question is deeper than a catalogue of cognitive errors: how much has already been decided by the way a problem reaches us?
Kahneman in five minutes
Daniel Kahneman (1934–2024) was a psychologist, born in Tel Aviv while his family lived in Paris. His childhood was marked by the persecution of Jews in occupied France. After studying in Jerusalem, he earned a doctorate in psychology at Berkeley in 1961. His academic life took him from the Hebrew University through Canada and California to Princeton.
He received the 2002 Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel for bringing psychological research into economics, especially the study of judgment and decision-making under uncertainty. The recognition rested substantially on research undertaken with Amos Tversky, who died in 1996. Kahneman was not an economist, nor the sole founder of behavioral economics.
His work extended from attention and effort to the psychology of well-being and organizational judgment. Thinking, Fast and Slow (2011) made much of it widely accessible; Noise (2021), written with Olivier Sibony and Cass Sunstein, turned to unwanted variation across decisions.
Before deciding, a mind must attend
Long before prospect theory, Kahneman studied attention as a limited cognitive resource. Attention and Effort (1973) asked how finite mental capacity is allocated. A decision cannot weigh every fact equally when some facts never enter awareness, while others consume attention because they are vivid or immediate.
During military personnel selection, Kahneman encountered a related difficulty: evaluators confidently inferred future performance from brief observations that had little predictive power. Structured assessments offered a more reliable alternative to the compelling but fragile overall impression. The episode illuminates a lifelong concern: subjective certainty is not a measurement of accuracy.
Heuristics: a mind working under constraints
In their 1974 paper, Tversky and Kahneman described three influential shortcuts in judgment under uncertainty. Availability uses the ease with which examples come to mind when estimating how common something is. Representativeness relies on resemblance to a familiar pattern, sometimes at the expense of base rates. Anchoring and adjustment can pull a numerical estimate towards a starting value.
These shortcuts make complex judgment possible without an exhaustive calculation. A heuristic is a procedure; a bias is a systematic deviation that may appear under particular conditions. Equating the two turns useful psychological research into a doctrine of universal human incompetence. Performance depends in part on how the task and its information are presented.
Prospect theory: where value begins
With Tversky, Kahneman developed prospect theory in 1979 as a descriptive account of choices under risk. People may evaluate an outcome as a gain or loss relative to a reference point rather than solely through their final wealth. The experienced difference between two nearby amounts can shrink as they move farther from that point, and decision weights need not follow probabilities linearly.
Losses may carry greater psychological weight than comparable gains in many settings. Loss aversion is not a universal two-to-one law. Its presence and magnitude vary with context and method. The theory was later extended through cumulative prospect theory (1992). It describes important patterns without supplying a formula that will predict every decision by every person.
Framing: the same outcome, another question
A description presents some aspects of a situation while leaving others in shadow. In their 1981 work on framing, Tversky and Kahneman found that logically equivalent outcomes could elicit different preferences when expressed in terms of gains rather than losses. The classic research on a hypothetical public-health emergency made this problem particularly visible.
The figure below uses a different, entirely invented example. It changes the wording around the same figures, then restores the context the figures cannot provide. The data need not change for the problem to look different. This is a possibility demonstrated by research on framing, not a claim that every reader will react in the same way.
A closer look · A problem before a preference
The numbers stay. The frame moves.
Imagine a hypothetical service that completes 90 of 100 requests on time. The result can be described in two equivalent ways. A third view asks what those figures still cannot tell us.
A reading device, not a behavioural test. These figures and this service are invented for the guide. The framing literature does not establish how any particular visitor will respond.
An original editorial illustration drawing on Tversky and Kahneman, “The Framing of Decisions and the Psychology of Choice” (1981), and the problem of incomplete evidence. All three views remain available in print.
Fast impressions, deliberate checks
The familiar labels System 1 and System 2 distinguish broadly automatic, rapid processes from more effortful and controlled ones. They are functional descriptions, not two anatomical parts of the brain. Slow thought can rationalize a poor answer, while trained intuition can be highly competent in environments with stable regularities and informative feedback.
Kahneman's shorthand WYSIATI — What You See Is All There Is — names the tendency to build a coherent account from the evidence at hand while giving too little weight to what is missing. An interviewer may answer the easier question "Does this person seem impressive?" without noticing that it has replaced "Will this person succeed in this role?"
The experience lived and the experience remembered
Another line of research distinguished the experiencing self from the remembering self. Retrospective judgments may be shaped especially by intense moments and by how an episode ended. In some experiments, duration received surprisingly little weight: the peak-end rule and duration neglect identify particular patterns, not a universal law of memory.
What was lived and what is remembered can thus lead to different evaluations of the same period. A recollection influences later choices and, when retold as a rating or recommendation, may become evidence for other people's judgments. This is one point of contact with Tarde: the memory that one person expresses can circulate and become a model for another.
When judgment becomes an institutional problem
A bias can move judgments consistently away from an appropriate standard. Noise is unwanted variability between assessments that ought to be comparable: two reviewers, for example, may reach sharply different conclusions about the same application under the same stated criteria.
In Noise (2021), Kahneman, Sibony and Sunstein argued for procedures that make assessments more consistent: separate the dimensions to be evaluated, collect independent judgments before discussion, record reasons and compare outcomes. Such practices cannot guarantee justice or accuracy, and uniformity is not itself a moral goal. They ask organizations to design conditions in which evidence can matter more than chance impressions.
The limits of a powerful research programme
The study of heuristics prompted important criticisms, notably from Gerd Gigerenzer and researchers of ecological rationality: a simple procedure may work very well when it matches its environment. Judgments about base rates also change when information is represented in natural frequencies or when experience provides useful feedback.
A different challenge concerned reproducibility. Some social-priming findings discussed in Thinking, Fast and Slow faced serious replication difficulties. In 2012 Kahneman himself urged researchers to conduct credible replications. It would be misleading to extend those doubts indiscriminately to every part of his work. A multinational study published in 2020 reproduced many foundational patterns of prospect theory, while identifying variation in effect strength. Each claim deserves its own evidence.
The contemporary wound: the prepared choice
A digital interface arranges a list before a person compares its items. A forecast displays a benchmark; a recommendation foregrounds one risk; an AI-generated explanation offers a fluent narrative whose missing evidence may be difficult to see. These are contemporary applications of questions Kahneman helps us ask, not technologies he predicted.
Some conditions of judgment are deliberately designed; others arise from ordinary limits of attention or from institutional noise. Making those conditions visible is therefore a question of governance as well as individual caution. Who chooses the frame, the reference point and the criteria by which a decision will later be called reasonable?
The triad now returns to its opening problem. Epicurus asks which desires deserve to govern a life. Tarde follows the beliefs and models through which desire circulates. Kahneman asks what happens when those influences must finally be judged. Their arguments remain distinct, but their tensions make freedom before choice a more exacting question.
Primary texts and orientation
Primary texts: Daniel Kahneman, Attention and Effort (1973); Amos Tversky and Daniel Kahneman, "Judgment under Uncertainty: Heuristics and Biases" (1974); Daniel Kahneman and Amos Tversky, "Prospect Theory: An Analysis of Decision under Risk" (1979); Amos Tversky and Daniel Kahneman, "The Framing of Decisions and the Psychology of Choice" (1981); "Advances in Prospect Theory" (1992); Thinking, Fast and Slow (2011); Daniel Kahneman, Olivier Sibony and Cass Sunstein, Noise (2021).
The ninth triad is now a complete English reading path: from Epicurus' measure of desire, through Tarde's social relations, to Kahneman's analysis of judgment.
