The Soldier Who Mapped the Mind: A Portrait of Amos Tversky

Opening: A room, suddenly his

On a winter afternoon in Jerusalem in the late 1960s, a dozen graduate students sat in Daniel Kahneman’s seminar on human judgment waiting, somewhat nervously, for class to begin. Kahneman almost never invited guests. When a compact paratrooper with close-cropped hair and a quick, mischievous smile slipped into the room and took an empty chair, the atmosphere shifted. Within minutes Amos Tversky was teasing the assumptions behind a student’s research proposal, inventing variants of the problem on the fly, and turning the conversation into a kind of high-speed collaborative game. By the end of the session the students were no longer quite sure whose seminar it was.[^1][^2][^3][^4]

That was a typical Tversky entrance. He had the bearing of an officer, the humor of a stand‑up comic, and the intellectual speed of someone who seemed to think in proofs and counterexamples. He could walk into almost any room in psychology or economics and, without trying, become its center of gravity. Yet the same man who delighted in dismantling arguments at the blackboard had once run toward an explosive charge under live fire to save a terrified conscript, an act of apparently reckless courage that left shrapnel in his body for the rest of his life.[^5][^6]

The puzzle of Amos Tversky is how these traits fit together: the decorated paratrooper who devoted his career to charting our irrational fears, the mathematically rigorous theorist who loved jokes, the dominant half of one of the great scientific partnerships of the twentieth century whose name is less widely known than his collaborator’s. To understand him is to see the field of judgment and decision making, and its history, from a slightly different map.[^6][^3][^5]

Haifa, Jerusalem, and the making of a mind

Amos Nathan Tversky was born in Haifa in 1937, when the city was still part of British Mandatory Palestine. His father, Yosef, was a Polish-born veterinarian who had originally trained in medicine, and his mother, Jenia (Genia) Ginzburg Tversky, was a Lithuanian Jewish social worker who became a member of the Knesset for the Mapai party from 1949 until her death in 1964. The household combined hard-headed practicality with a deep engagement in politics and social welfare; arguments were a way of life rather than a sign of conflict.[^4][^5][^6]

Jenia later said that Amos was "self-taught in many areas, including mathematics," a remark that fits with accounts of a boy who devoured books far outside his formal curriculum. In high school he studied with the literary critic Baruch Kurzweil and befriended Dahlia Ravikovich, who would become one of Israel’s major poets. The pairing is suggestive: from early on his intellectual world mixed formal structure and metaphor, rigorous argument and sensitivity to how language shaped experience.[^5][^6]

The new state of Israel was itself an education in uncertainty. Tversky’s parents had arrived from Eastern Europe by way of Russia, and the family’s story carried the memory of displacement, risk, and political commitment. In that context, the young Amos gravitated toward both philosophy and psychology at the Hebrew University of Jerusalem, where he earned his bachelor’s degree in 1961, majoring in both fields. Even as an undergraduate he already had, in the words of one biographical account, a "clear vision" that he wanted to study judgment.[^6][^4][^5]

The soldier: three wars and a grenade

Like most Israelis of his generation, Tversky’s intellectual life ran in parallel with compulsory military service. After high school he was drafted into the Israel Defense Forces and joined the Paratroopers Brigade, making more than fifty jumps and eventually rising to the rank of captain. He also served in Nahal, an IDF program that combined military service with the establishment of kibbutzim and agricultural settlements, a blend of idealism and toughness that marked his generation of officers.[^5][^6]

The defining story of his military career took place not in combat but during a training exercise in 1956, in front of the IDF General Staff. As a platoon commander, he ordered a soldier to crawl forward and place a Bangalore torpedo under a barbed wire obstacle, light the fuse, and then run back to cover. The soldier lit the fuse, froze in panic, and stayed beside the live charge while the rest of the unit sheltered behind a rock. Disregarding orders to stay down, Tversky ran out, grabbed the soldier, and dragged him away just before the explosion. He was wounded by shrapnel, which remained in his body for the rest of his life, but the soldier escaped unharmed.[^6][^5]

Accounts agree that Moshe Dayan, then chief of staff, witnessed the event and told Tversky when awarding him Israel’s highest military decoration that he had done something "very stupid and very brave" and should not expect to get away with it again. The comment captures a tension that would later animate Tversky’s scientific work: courage may involve a deliberate violation of expected-utility calculations, a choice to privilege a person, a principle, or a narrative over statistical self-preservation.[^5][^6]

Tversky fought in three wars: he parachuted into combat during the 1956 Suez Crisis, commanded an infantry unit in the Six-Day War in 1967, and served in a psychology field unit during the Yom Kippur War in 1973. Colleagues later remarked that the man who mapped systematic biases in risk perception was personally almost indifferent to physical danger, and that he carried himself in academic life with the calm, direct manner of an officer used to command.[^7][^4][^6][^5]

The mathematician of judgment: early work before Kahneman

Tversky’s formal training in judgment began at the University of Michigan, where he arrived on a Fulbright scholarship for doctoral study in mathematical psychology and received his PhD in 1965. There he worked with Clyde Coombs, founder of the mathematical psychology program, and met fellow graduate student Barbara Gans, whom he married in 1963. Barbara Tversky would later become a distinguished cognitive psychologist in her own right; together they raised three children, Oren, Tal, and Dona.[^8][^6][^5]

His dissertation combined two strands that would run through his career: rigorous axiomatic analysis of choice and an experimental challenge to expected utility theory. It introduced conditions under which people’s preferences among risky prospects could become systematically intransitive, violating the foundational axiom that if you prefer A to B and B to C, you must prefer A to C. In a 1969 Psychological Review paper, "Intransitivity of Preferences," he showed that under carefully designed conditions people produce consistent, predictable cycles in their choices. Rather than treating this as noise, he argued that it revealed the simplifying procedures people use when trading off attributes, and the ways those procedures can generate structured violations of rational-choice axioms.[^9][^10][^6]

During this period he also became a central figure in the emerging foundations-of-measurement program. Working with David Krantz, R. Duncan Luce, and Patrick Suppes, he coauthored the three-volume treatise Foundations of Measurement, which provided axiomatic conditions for representing qualitative preference and similarity judgments on numerical scales. This work rarely appears in popular accounts of behavioral economics, yet within cognitive science it established Tversky as one of the leading formal theorists of his generation.[^6]

Perhaps his most celebrated solo contribution is the 1977 paper "Features of Similarity" in Psychological Review. There he challenged the then-dominant view that similarity could be modeled as distance in a geometric space with symmetric, metric properties. Using simple but revealing experiments, he showed that people’s similarity judgments are often asymmetric: a son is easily said to resemble his father, but we rarely say that a father resembles his son; Tel Aviv is judged more similar to New York than New York is to Tel Aviv.[^11][^12][^6]

To account for these anomalies, Tversky proposed the "contrast model" of similarity. Objects are represented as sets of qualitative features; the similarity of A to B is a weighted combination of the features they share and the features distinctive to each. Asymmetry arises when the more "prominent" object contributes more distinctive features, or when attention is directed to one item as the subject and the other as the referent. This seemingly technical shift, from distances in a metric space to feature-based contrasts, anticipated later developments in cognitive science and machine learning that represent concepts as high-dimensional feature vectors rather than points in a simple Euclidean geometry.[^13][^12][^11]

Before he ever published with Daniel Kahneman, in other words, Tversky had already reshaped theories of measurement, choice, and similarity. His solo work would alone justify a place in the canon of cognitive psychology.

The partnership: what Amos brought to Kahneman

The collaboration with Kahneman began in the late 1960s in Jerusalem, when both were on the faculty of the Hebrew University. By Kahneman’s own account, their first serious conversation was a kind of mutual confession of statistical sins: each recounted mistakes they had made in thinking about data and probability, and they resolved to study the "statistical intuitions of experts" together. What followed was a decade-long near-mind‑meld in which, as Michael Lewis later put it, the two men "completed each other’s sentences" and could no longer easily say which ideas belonged to whom.[^14][^15][^16][^17]

Kahneman’s 2003 essay "Experiences of Collaborative Research" offers the most authoritative glimpse into the division of labor. He emphasizes that they avoided explicit role-splitting, working through almost every line of every paper together in the same room, arguing until both were satisfied. Still, he notes that they each brought a different sensibility: Kahneman was more attuned to the phenomenology of judgment, the texture of what it feels like to make a prediction or a choice, whereas Tversky had what he calls "simply perfect taste in choosing problems" and an unfailing ability to formalize an intuition into a clean experimental design or a precise model.[^16][^14][^6]

Colleagues often described Tversky as the mathematician of the pair, the one who could turn a verbal hunch into a theorem or a counterexample. That characterization risks oversimplifying both men, but it captures something real about Amos’s contribution: he insisted that qualitative insights into bias be embedded in rigorous structures, whether in the axioms of measurement, the value function of prospect theory, or the contrast model of similarity.[^18][^6]

He also brought a particular style of intellectual combat. Kahneman later remarked that in argument Tversky "wanted to crush the opposition," and that this ruthless clarity helped them strip away sloppy thinking, including their own. In the context of their partnership, that aggression was largely turned inward, protecting the work from error rather than winning points in public debates.[^5][^6]

The papers that changed everything

Heuristics and biases

The first wave of their joint work focused on "judgment under uncertainty" and the shortcuts, or heuristics, people use when assessing probabilities and frequencies. In their 1974 Science article, "Judgment under Uncertainty: Heuristics and Biases," they distilled this research into three core heuristics: representativeness, availability, and anchoring and adjustment.[^19][^20]

From Tversky’s angle, these were not vague metaphors but specific, testable procedures. Representativeness says that people judge the likelihood that A belongs to B by the degree to which A resembles a typical member of B, often ignoring base rates and sample sizes. Availability says that people estimate the frequency or probability of an event by how easily instances or scenarios come to mind; events that are more salient, vivid, or recent are judged more probable, regardless of actual frequencies. Anchoring and adjustment describes how numerical estimates often start from an initial value, which may be arbitrary, and then adjust insufficiently away from it, leaving final judgments systematically biased toward the anchor.[^21][^20][^19]

The Science article was a synthesis rather than a technical advance, but its influence was catalytic. It provided economists, lawyers, policy analysts, and statisticians with a simple vocabulary for talking about systematic departures from rational models, and it showcased Tversky’s hallmark style: deceptively simple experimental vignettes that induced striking, replicable deviations from normative probability theory.[^22][^6]

Prospect theory and the value of losses

If the heuristics-and-biases work showed that people violated the laws of probability, prospect theory showed that they also violated expected utility theory, the foundation of orthodox models of choice under risk. In their 1979 Econometrica paper "Prospect Theory: An Analysis of Decision under Risk," Tversky and Kahneman proposed an alternative descriptive model.[^23][^24]

The key moves were reference dependence, loss aversion, diminishing sensitivity, and nonlinear probability weighting. Outcomes are evaluated as gains or losses relative to a reference point, typically the status quo, rather than in terms of final wealth. The value function is concave for gains and convex for losses, with a steeper slope for losses, capturing the empirical fact that "losses loom larger than gains." And instead of using objective probabilities as weights, the theory uses decision weights that overweight small probabilities and underweight moderate to large ones.[^25][^26]

Tversky’s mathematical fingerprints are especially visible in the shape of the value function and the probability weighting function. Later work on "cumulative prospect theory" refined the weighting into a rank-dependent form and generalized the model to a wider class of gambles, a development that bears his name as coauthor. Together with the 1991 paper "Loss Aversion in Riskless Choice: A Reference-Dependent Model," in which Tversky and Kahneman showed how loss aversion and reference points could reshape indifference curves in standard consumer theory, this body of work anchored the now-dominant idea that preferences are defined over changes, not states, and that negative changes count more.[^27][^28][^29][^18][^6]

Richard Thaler’s program on mental accounting and the endowment effect can be read as a sustained elaboration of the reference-dependent framework Tversky helped build. Thaler explicitly cites Kahneman and Tversky’s notion of a "mental account" and their theater-ticket examples as foundations for his own account of how people code gains and losses in psychologically distinct buckets rather than in a single global wealth measure. Tversky did not write the canonical papers on mental accounting, but his conceptual role in defining reference dependence and loss aversion is visible throughout that literature.[^30][^31][^32][^27]

Framing: the Asian Disease and the fragility of rational choice

If prospect theory cracked the edifice of expected utility, the 1981 Science paper "The Framing of Decisions and the Psychology of Choice" undermined the presumption that rational choice is invariant under merely verbal transformations. Tversky and Kahneman showed that people’s preferences between formally equivalent options could be reversed by framing outcomes as gains or losses, or by changing which outcomes were described as certain versus probabilistic.[^33][^34][^35]

The most famous example is the "Asian Disease" problem. Subjects were told that an unusual disease was expected to kill 600 people and asked to choose between two programs. In a gain frame, Program A would "save 200 lives," while Program B offered a one-third chance of saving all 600 and a two-thirds chance of saving none; most participants chose the sure saving of 200. In a loss frame, Program C meant that "400 people will die," while Program D offered a one-third chance that no one would die and a two-thirds chance that all 600 would die; now most participants preferred the risky option.[^36][^35][^33][^4]

Because the two pairs of options are mathematically equivalent, a rational decision maker with stable preferences should not reverse choices simply because outcomes are described in terms of lives saved versus lives lost. Yet the majority did, in systematic ways that align with prospect theory’s prediction of risk aversion over gains and risk seeking over losses. For rational choice theorists, this was not a minor anomaly but a direct challenge to the invariance principle, the idea that preferences depend only on the substantive content of outcomes, not their description.[^34][^35]

The conjunction fallacy and Linda

In "Extensional Versus Intuitive Reasoning: The Conjunction Fallacy in Probability Judgment" (1983), Tversky and Kahneman attacked another pillar of rational inference: the rule that the probability of a conjunction, P(A and B), cannot exceed the probability of either conjunct alone. The now-classic Linda problem asked participants to read a description of Linda as a 31‑year‑old, single, outspoken, philosophy major deeply concerned with social justice and antinuclear activism, then rank various statements about her by their probability.[^37][^38][^4]

Two critical statements were: (1) "Linda is a bank teller" and (2) "Linda is a bank teller and is active in the feminist movement." Across samples, about 85–90 percent of respondents judged the conjunction more probable than the simpler statement, a direct violation of the conjunction rule. Tversky and Kahneman argued that people were relying on representativeness rather than extension: the detailed, stereotype-consistent description made Linda seem more like a feminist bank teller, even though the set of feminist bank tellers is necessarily a subset of all bank tellers.[^39][^38][^37]

The Linda problem became a lightning rod in debates about human rationality, Bayesianism, and the proper interpretation of ordinary language. For Tversky, it was another instance where intuitive reasoning followed a coherent but non-extensional logic, one guided by similarity and narrative coherence rather than by set inclusion. The conjunction fallacy also provided a proving ground for later models of subjective probability, including Tversky’s own support theory.[^40][^38]

The hot hand: seeing streaks in randomness

One of Tversky’s most culturally visible works, done with Thomas Gilovich and Robert Vallone, took place on the basketball court rather than in the laboratory. In "The Hot Hand in Basketball: On the Misperception of Random Sequences" (1985), they examined the widespread belief among fans and players that shooters go through "hot" streaks during which they are more likely to make subsequent shots.[^41][^42]

Analyzing field-goal data from the Philadelphia 76ers and free-throw sequences from the Boston Celtics, as well as controlled shooting in the gym, they found little evidence that success on one shot increased the probability of success on the next; the patterns in hits and misses looked like what one would expect from independent Bernoulli trials, given a player’s overall shooting percentage. Yet surveys showed that most fans and players believed strongly in the hot hand and adjusted their in-game decisions accordingly, for example by feeding the "hot" shooter.[^42][^41]

The paper argued that people are poor at intuiting the properties of random sequences, overinterpreting inevitable clusters of success as evidence of streaky performance. Later reanalyses by Joshua Miller and Adam Sanjurjo, among others, showed that the original statistical tests underestimated the prevalence of genuine hot-hand effects because of subtle biases in the estimators, and that modest positive dependence in shooting was consistent with the data. Still, even these critics affirmed the psychological point: people tend to overinfer streaks from limited data, a pattern squarely in line with the earlier work on representativeness and the "law of small numbers."[^43][^44]

Support theory: unpacking and the weight of description

Tversky’s last major theoretical contribution, developed with Derek Koehler and elaborated with Yuval Rottenstreich, was "support theory," a nonextensional account of subjective probability. Starting from the observation that judged probabilities assigned to the components of an event (for example, specific accident causes) often sum to more than the judged probability of the inclusive event (accidental death), they proposed that people assign probability in proportion to the "support" that a description evokes, not to its extension in a formal sample space.[^45][^46][^47][^48]

In this framework, unpacking an event into explicit, vivid subcases tends to increase its total judged probability (implicit subadditivity), while repacking can reduce it, and explicit disjunctions may receive less probability than the sum of their parts (explicit subadditivity). The same objective event, described in a sparse or rich way, can therefore receive different probability judgments, integrating and generalizing many of the heuristics-and-biases phenomena into a single formalism. Support theory closed a kind of circle: the young Tversky who had shown that similarity depends on feature representations now offered a parallel account of how probability depends on the descriptive richness of hypotheses.[^46][^47]

Stanford and the later years

After teaching at Hebrew University through the 1960s and early 1970s and spending periods at the Oregon Research Institute and Harvard, Tversky moved permanently to Stanford University in 1978. There he became the first Davis‑Brack Professor of Behavioral Science, helped found the Stanford Center on Conflict and Negotiation, and became a central figure not only in psychology but in an emerging interdisciplinary network that included economists, philosophers, statisticians, and legal scholars.[^49][^7][^6]

By all accounts he was a dominant presence in the department. Stanford’s memorial resolution, as summarized in Encyclopedia.com and other sources, describes him as a leading decision researcher whose work challenged economic theory and influenced law, medicine, political science, and statistics. Colleagues recalled that one could design a new experiment in a single lunchtime conversation with him, and that he would often continue working alone late into the night after a day of collaborative meetings.[^7][^6][^5]

His personality combined warmth and ferocity. Persi Diaconis, a Stanford mathematician, said that "there was a light shining out of him," while Kahneman later called him "the freest person" he had known, simultaneously highly disciplined and almost unconcerned with status. At conferences he was, as one reviewer of The Undoing Project put it, "the guy everyone wanted to talk to at a party"; stories circulated about the "Tversky intelligence test," which held, half-jokingly, that the faster you realized he was smarter than you, the smarter you were.[^15][^4][^6][^5]

Within the field, Tversky became a kind of north star. David Laibson and Richard Zeckhauser titled a long appreciation "Amos Tversky and the Ascent of Behavioral Economics," tracing how his experiments and models turned behavioral critiques of rationality from a marginal pursuit into a central agenda in economics. His MacArthur Fellowship citation in 1984 emphasized his "extraordinary originality" and the way his work on "cognitive illusions" had reshaped understanding of decision making across disciplines.[^50][^49][^6]

The diagnosis and the silence

In early 1996, at the age of fifty‑eight, Tversky was told he had metastatic melanoma and only months to live. Rather than make a public announcement, he chose to keep the diagnosis within a tight circle of family and close colleagues. According to biographical accounts, he continued working almost to the end, editing with Kahneman the anthology Choices, Values, and Frames and leaving a dozen papers in press at the time of his death on June 2, 1996, at his home in Stanford.[^7][^6]

The decision to maintain privacy about his illness fits with the way he had handled military danger and intellectual conflict: the focus remained on the work, not on his vulnerability. Friends later recalled that he talked more in those months about unfinished ideas than about his own mortality. For a field that had spent decades studying how people frame losses and confront risk, there is something fitting and piercing in the image of its leading theorist calmly editing proofs under a death sentence he largely chose not to share.

The Nobel that never was

Six years after Tversky’s death, in 2002, the Royal Swedish Academy of Sciences awarded the Nobel Memorial Prize in Economic Sciences to Daniel Kahneman, shared with experimental economist Vernon Smith, "for having integrated insights from psychological research into economic science." Because Nobel Prizes are not given posthumously, Amos Tversky could not be named as a co‑laureate.[^51][^50]

Kahneman used his Nobel lecture, "Maps of Bounded Rationality," to underscore that "the work cited by the Nobel committee was done jointly with the late Amos Tversky (1937–1996) during a long and unusually close collaboration." He went on to frame the lecture as a review of "our joint work" on heuristics, prospect theory, and framing, repeatedly marking Tversky’s contributions by name. The Encyclopedia.com biography notes that he began the talk by explicitly crediting Amos and that he regarded the award as a "joint prize" in spirit.[^52][^53][^50][^6]

Outside observers recognized the structural injustice of the situation. A Chicago Tribune column in June 1996, titled "Great Men, Great Prize, Death Denies," paired Tversky with Fischer Black as two scholars whose early deaths likely cost them a Nobel, and lamented the prize’s unwillingness to acknowledge such posthumous debts. Later essays and obituaries similarly stressed that had Tversky lived a few more years he would almost certainly have shared the honor.[^54][^4][^6]

As for Tversky’s own attitude, The Undoing Project and related profiles suggest that in the early 1990s he was more aware than Kahneman of how their work had been received in economics journals and how prize committees tended to think, but there is little evidence that he obsessed over it. Colleagues describe him as ambitious for the ideas rather than the awards; the injustice is felt more acutely by the field than, so far as the record shows, it was by Amos himself.[^3][^17][^1]

What was lost at fifty‑nine

What, concretely, was lost when Tversky died at fifty‑nine? One answer is simply: continuation. At the time of his death he was actively developing support theory, refining reference-dependent models of choice, and collaborating with younger scholars on reason-based decision making, such as the "disjunction effect" experiments with Eldar Shafir. His intellectual trajectory pointed toward increasingly general frameworks that could integrate heuristics, framing, and subjective probability into unified accounts of how descriptions shape judgment.[^47][^46][^6]

Encyclopedia.com notes that he left twelve articles in press and that many of his existing papers had already become canonical in multiple fields. Given his history of returning to old questions with new tools, it is easy to imagine him extending prospect theory to dynamic settings, clarifying the interplay between affect and cognition in risk perception, or engaging more deeply with game theory and negotiation through the Stanford Center on Conflict and Negotiation.[^6]

More speculatively, his presence might have altered the second-generation debates over behavioral economics. As the field moved from identifying biases to designing "nudges" and choice architectures, there was a risk of treating the catalog of heuristics as a menu of levers for policymakers rather than as windows into the structure of cognition. Tversky’s insistence on formal clarity and his discomfort with sloppy applications suggest he would have been a powerful voice reminding economists that the point was not simply to exploit biases, but to understand them.

Why Kahneman is famous and Tversky is not (quite)

In the popular imagination, Daniel Kahneman is the face of judgment and decision making, thanks largely to his 2011 book Thinking, Fast and Slow, which distilled decades of research into an accessible narrative and sold millions of copies. Amos Tversky, by contrast, left no memoir, no general-audience book, and no Nobel lecture under his own name. His reputation rests in technical papers and in the recollections of colleagues, plus Michael Lewis’s The Undoing Project, which is as much a portrait of their relationship as of either man individually.[^55][^56][^17][^15][^51][^5][^6]

Several structural factors help explain the asymmetry. Kahneman outlived Tversky by nearly three decades and had time both to consolidate their scientific legacy and to interpret it for broader audiences. The Nobel Prize, whatever Kahneman’s own framing, inevitably attached his name more strongly to the work in public discourse. And personality mattered: Kahneman’s self-critical, confessional voice lent itself to narratives of "how I discovered my own irrationality," while Tversky’s brisk, high‑confidence style was less easily translated into the first-person essays and talks that drive popular fame.[^57][^14][^1][^51][^5]

Yet the idea that Tversky was the "supporting character" in Kahneman’s story is hard to square with the historical and bibliographic record. He published more articles in Psychological Review than any other psychologist of the journal’s first century, coauthored the foundational treatise on measurement, and appears as lead or co‑lead author on nearly all of the landmark papers that undergird behavioral economics. Kahneman himself repeatedly insisted that the prize‑winning work was joint, that Amos had "perfect taste in problems," and that he regarded their contributions as inextricable.[^58][^14][^52][^6]

For the field, then, "catching up" with history means learning to see the map of bounded rationality as an Amos-and-Danny construction rather than a Danny-with-Amos footnote, and to appreciate the distinctively Tverskyan elements: the contrast models, the axioms, the insistence that cognitive illusions be derived from explicit assumptions rather than loose metaphors.

A solo legacy, reconstructed

If one could, counterfactually, strip away the Kahneman collaboration and look only at Tversky’s solo or non-Kahneman work, what legacy would remain? The answer is: a remarkably rich one.

First, there is the measurement and choice theory: "Intransitivity of Preferences," the conjoint measurement work, elimination by aspects, and preference trees. These contributions reshaped how psychologists and economists model trade‑offs among attributes and how they think about the conditions under which transitivity can fail in systematic ways.[^10][^9][^6]

Second, there is the theory of similarity. "Features of Similarity" remains one of the most cited and discussed papers in cognitive psychology, in part because it connected philosophical questions about resemblance to empirical anomalies and offered a simple, powerful mathematical model. Later work has shown how versions of the contrast model can be embedded in Bayesian and connectionist frameworks, but the core intuition—that similarity judgments depend on feature salience and direction, not just distance—remains foundational.[^12][^13][^11]

Third, there is the late work on support theory and reason‑based choice. Even setting aside Kahneman, Tversky’s collaborations with Derek Koehler, Yuval Rottenstreich, Eldar Shafir, and others laid out a set of phenomena—unpacking, superadditivity, the disjunction effect—that any comprehensive theory of subjective probability and choice must now explain.[^59][^46][^47]

Finally, there is the thematic thread that runs through all of it: a view of people as intuitive scientists who are very good at generating hypotheses and patterns but "not very good at testing hypotheses," as Tversky once put it. His models treat heuristics not as flaws to be mocked but as sensible procedures in a complex, probabilistic world, procedures that become biased only when pushed beyond their domain or when misled by how options and events are described.[^4]

Viewed this way, Tversky’s solo intellectual profile is that of a theorist of representation and comparison, someone who repeatedly showed how the mind uses sparse, feature-based codes to handle overloaded problems of choice, similarity, and probability. The Kahneman partnership amplified and popularized that vision, but it did not create it from scratch.

The soldier who knew his work would outlast him

Near the end of The Undoing Project, Michael Lewis quotes friends who said that Tversky seemed almost constitutionally uninterested in regret. Having faced live explosives, three wars, and the uncertainties of building a new field, he had little patience for counterfactual brooding about prizes or missed opportunities. What mattered was whether an idea was elegant, whether an experiment really nailed a phenomenon, whether a model cut nature at the joints.[^17][^1]

The irony is that few scientists have done more to show how powerfully counterfactuals shape human experience. Prospect theory, framing effects, support theory, and the disjunction effect all reveal minds that live as much in imagined alternatives—the lives saved versus lives lost, the paths taken versus untaken—as in the outcomes that actually occur. Tversky’s soldier’s stoicism sat alongside an intellectual career devoted to mapping exactly how fragile that stoicism can be.[^35][^25][^46]

He died in 1996, in a small house in Stanford, with shrapnel still in his body from a training field in 1956 and with unfinished manuscripts on his desk. Six years later his closest collaborator stood in Stockholm and accepted a Nobel Prize for work they had done together, insisting that it was a joint achievement and offering, in the title "Maps of Bounded Rationality," a phrase that could just as well describe Amos Tversky’s life project.[^50][^52][^7][^5][^6]

To read his papers today is to enter a mind that delighted in finding the precise point where common sense and formal rationality part ways, and then in building a bridge between them. The maps he drew of our bounded rationality, our similarity judgments, our susceptibility to framing and unpacking, have become fixtures in psychology, economics, law, and beyond. What has been catching up, slowly, is recognition of the soldier who drew them.

---

References

1. A Meeting of Two Minds - Public Books - We all make mistakes. No matter how well trained, smart, or astute we are, faulty reasoning marks ou...

2. HOW TWO TRAILBLAZING PSYCHOLOGISTS TURNED THE WORLD OF DECISION SCIENCE UPSIDE DOWN - Vanity Fair:The dozen or so graduate students in Danny Kahneman’s seminar at Hebrew University, in J...

3. The Two Friends Who Changed How We Think About How We Think - The New Yorker · Page-Turner. The Two Friends Who Changed How We Think About How We Think. By Cass R...

4. Behavioral economics and law

5. Amos Tversky - Wikipedia

6. Tversky, Amos - Encyclopedia.com - TVERSKY, AMOS(b. Haifa, Israel, 16 March 1937;d. Stanford, California, 2 June 1996), judgment and de...

7. Amos Tversky, leading decision researcher, dies at 59

8. [PDF] July 2016 - The Regents of the University of Michigan - Ann Arbor - Amos Tversky was born in Haifa, Israel in 1937. He received his Bachelor of Arts from Hebrew. Univer...

9. [PDF] Intransitivity of preferences. | Semantic Scholar - It is shown that, under specified experimental conditions, consistent and pre- dictable intransitivi...

10. INTRANSITIVITY OF PREFERENCES

11. [PDF] Features of Similarity - Semantic Scholar - Features of Similarity. @article{Tversky1977FeaturesOS, title={Features of Similarity}, author={Amos...

12. [PDF] 1Features of Similarity - '' The. Features of Similarity. 17. Page 12. lists were constructed so that the more prominent count...

13. Using Tversky's contrast model to investigate how features of ... - Using Tversky's contrast model to investigate how features of similarity affect judgments of likelih...

14. [PDF] Experiences of Collaborative Research - We conducted a series of market experiments involving real goods (the. “mugs studies”), which eventu...

15. The Undoing Project - Jewish Book Council - The focus of this book is not a person, or even two people, but the relationship between two extraor...

16. EJ775360 - Experiences of Collaborative Research, American ... - The author's personal history of the research that led to his recognition in economics is described,...

17. The undoing project : a friendship that changed our minds

18. Prospect Theory: An Analysis of Decision under Risk - IDEAS/RePEc - This paper presents a critique of expected utility theory as a descriptive model of decision making ...

19. Judgment under Uncertainty: Heuristics and Biases

20. Judgment under Uncertainty: Heuristics and Biases: Biases in judgments reveal some heuristics of thinking under uncertainty. - This article described three heuristics that are employed in making judgments under uncertainty: (i)...

21. Judgment under Uncertainty: Heuristics and Biases Author(s)

22. [PDF] Judgment under uncertainty: Heuristics and biases

23. Prospect Theory: An Analysis of Decision Under Risk - Analysis of decision making under risk has been dominated by expected utility theory, which generall...

24. [PDF] Amos Tversky Econometrica, Vol. 47, No. 2. (Mar., 1979), pp

25. Prospect Theory: An Analysis of Decision under Risk, Econometrica ... - From the document: "This paper presents a critique of expected utility theory as a descriptive ...

26. Prospect Theory - Center for Public Interest Communications - Prospect theory comes from behavioral economics and is credited to Daniel Kahneman and Amos Tversky ...

27. Loss Aversion in Riskless Choice: A Reference-Dependent Model. - Much experimental evidence indicates that choice depends on the status quo or reference level: chang...

28. Loss Aversion in Riskless Choice: A Reference-Dependent Model

29. [PDF] Loss Aversion in Riskless Choice: A Reference-Dependent Model

30. [PDF] Mental Accounting Matters - Mental accounting is the set of cognitive operations used by individuals and households to organize,...

31. TESTING REFERENCE DEPENDENCE, LOSS AVERSION AND ...

32. [PDF] Mental Accounting Matters - Andrei Simonov - Mental accounting is the set of cognitive operations used by individuals and households to organize,...

33. [PDF] The Framing of Decisions and the Psychology of Choice - In ef- fect, the respondents were asked to choose between 10 percent chance of losing 75 lives and 8...

34. The framing of decisions and the psychology of choice - PubMed - The framing of decisions and the psychology of choice. Science. 1981 Jan 30;211(4481):453-8. doi: 10...

35. The Framing of Decisions and the Psychology of Choice - Science - The Framing of Decisions and the Psychology of Choice. Amos Tversky and Daniel KahnemanAuthors Info ...

36. The Framing of Decisions and the Psychology of Choice. - ERIC - Publication Date: 1981-Jan ... The Framing of Decisions and the Psychology of Choice. Tversky, Amos;...

37. Extensional versus Intuitive Reasoning: The Conjunction Fallacy in ... - Extensional versus Intuitive Reasoning: The Conjunction Fallacy in Probability Judgment. Tversky, Am...

38. [PDF] Extensional Versus Intuitive Reasoning: The Conjunction Fallacy in ... - Extensional versus intuitive reasoning: the conjunction fallacy in probability judgment by Tversky, ...

39. Extensional versus intuitive reasoning: The conjunction fallacy in ... - Extensional versus intuitive reasoning: The conjunction fallacy in probability judgment ... 1983 by ...

40. 6 - Extensional Versus Intuitive Reasoning: The Conjunction Fallacy ... - 6 - Extensional Versus Intuitive Reasoning: The Conjunction Fallacy in Probability Judgment. Publish...

41. Gilovich (1985) The hot hand in basketball. On the misperception of random sequences

42. Hot hand - Wikipedia

43. 133 thoughts on “ Gilovich doubles down on hot hand denial ”

44. [88] The Hot-Hand Artifact for Dummies & Behavioral ... - A friend recently asked for my take on the Miller and Sanjurjo's (2018; pdf) debunking of the hot ha...

45. [PDF] A Foundation for Support Theory Based on a Non-Boolean Event ... - Tversky and Koehler (1994) and Rottenstreich and Tversky (1997) ... The relation between probability...

46. Unpacking, repacking, and anchoring: advances in support theory - PubMed - Support theory represents probability judgment in terms of the support, or strength of evidence, of ...

47. Strength of evidence, judged probability, and choice under uncertainty - The investigation relies on the theoretical framework provided by support theory (Tversky & Koehler,...

48. Amos Tversky & Derek J. Koehler, Support theory: A nonextensional ... - Support theory: A nonextensional representation of subjective probability · Amos Tversky & Derek J. ...

49. Amos Tversky - MacArthur Foundation - The MacArthur Foundation supports creative people and effective institutions committed to building a...

50. Maps of Bounded Rationality - IDEAS/RePEc - "Maps of Bounded Rationality," Nobel Prize in Economics documents 2002-4 ... Amos Tversky & Daniel K...

51. Maps of bounded rationality by Daniel Kahneman - Goodreads - Maps of bounded rationality. Daniel Kahneman. 3.82. 136 ratings ... He was awarded the 2002 Nobel Pr...

52. [PDF] Maps of Bounded Rationality: Psychology for Behavioral Economics† - KAHNEMAN: MAPS OF BOUNDED RATIONALITY. Page 20. Tversky, 1979). The normal ... : A Moment-Based Appr...

53. [PDF] Maps of Bounded Rationality: Psychology for Behavioral Economics - KAHNEMAN: MAPS OF BOUNDED RATIONALITY. Page 21. Tversky, 1979). The normal ... : A Moment-Based Appr...

54. GREAT MEN, GREAT PRIZE, DEATH DENIES - Chicago Tribune - GREAT MEN, GREAT PRIZE, DEATH DENIES. Share this: Share on Facebook (Opens in new window) Facebook ....

55. The Undoing Project - Wikipedia

56. Daniel Kahneman - The Decision Lab - Kahneman came to this idea after years of behavioral science experiments he conducted with Amos Tver...

57. Why I'll never forget the day I met Daniel Kahneman for lunch - Psyche - The work for which he won his Nobel Prize was conducted largely alongside his research partner, the ...

58. Amos Tversky biography, quotes and books - Toolshero - Amos Tversky was an Israeli psychologist and one of the most influential figures in the discovery of...

59. TVERSKY'S Intransitivity of Preference Revisited - Author(s): Regenwetter, Michel | Abstract: Transitivity of preference is defined as follows: if a pe...