here are, Secretary of Defense Donald Rumsfeld explained in the run-up to the Iraq War, “known knowns; there are things we know we know.” And, “there are known unknowns…we know there are some things we do not know. But there are also unknown unknowns—the ones we don’t know we don’t know.”

Rumsfeld’s statement, now famous and often derided, expressed a truth applicable beyond geopolitics: breakthroughs in the fields of behavioral economics and experimental psychology reveal just how badly human knowledge and intuition serve us in making decisions. In The Undoing Project, Michael Lewis explores advances in “the way the human mind work[s], or fail[s] to work, when…forming judgments and making decisions.” Undoing explicates the work of Daniel Kahneman and the late Amos Tversky, two psychologists at the Hebrew University of Jerusalem who forged an unlikely but extraordinarily fruitful partnership that deeply enriched our understanding of ourselves.

In many ways, the two Israelis were an unlikely couple. Kahneman immigrated to Israel after hiding with his family in occupied France during World War II, and typically found himself plagued by doubt, overly earnest, eager to please, and more than a bit anti-social. By contrast, Tversky, a native Israeli sabra, was jovial, confident, informal, and optimistic. And yet, as Lewis puts it:

Both were explicitly interested in how people functioned when they were in a “normal” unemotional state. Both wanted…to do “the psychology of simple questions”…to cut through endless bullshit to the simple nub of any matter. Both men were blessed with shockingly fertile minds. And both were Jews, in Israel, who did not believe in God.

They collided when Tversky delivered a captivating guest lecture in Kahneman’s Hebrew University seminar. As the latter vigorously debated the former, “something shifted inside Amos. He left Danny’s seminar in a state of mind unusual for him: doubt.” For his part, Kahneman “had never thought much about thinking” before his exposure to Tversky’s heterodox theories.

The pair undertook to investigate the deeply imperfect process of human decision-making. What they found wasn’t so much that people made mistakes—the knownest of knowns—as that “the mistakes were predictable and systematic.” Kahneman and Tversky’s discoveries have influenced fields as diverse as medicine, public health, economics, military preparedness and tactical training, aviation, sports, advertising, automotive safety, consumer choice, and the homelessness problem.

Their works’ abstractions were rooted in specific, real-world experimentation. Lewis, whose 2003 bestseller Moneyball described and catalyzed a revolution in player evaluation by baseball executives, proffers the case study of Daryl Morey, the whiz-kid general manager of the NBA’s Houston Rockets, who perhaps unwittingly deployed sophisticated decision-making techniques in weighing whom to draft or trade and how much to pay. Morey’s Moneyball-style approach initially met stiff resistance from generations of basketball hands who’d relied on gut, look, and feel, but eventually won wide acceptance.

“It was strange,” Lewis imagines Morey imagining:

that when people bothered to measure what happened on the court, they had measured the wrong things so happily for so long. It was bizarre that it was even possible for a total outsider to walk into the game with an entirely new approach to valuing basketball players and see his approach adopted by much of the industry.

For Lewis, the Morey story provides more than just an interesting anecdote. In fact, the scourges against which Morey battled mirror some of the most common errors in human judgment, and formed the basis of Tversky and Kahneman’s groundbreaking work.

“Representativeness,” for example: the comparison of whatever we’re evaluating to a mental model of the same thing. The Rockets’ scouts observed potential draftees practicing and automatically compared them to professional players with similar racial and body-type characteristics—a proclivity that Morey struggled to suppress. Kahneman and Tversky encountered this tendency when survey respondents chose which pattern of gender and birth-order was more common: {boy, boy, boy, girl, girl, girl}, or {girl, boy, boy, girl, boy, girl}. An overwhelming majority believed the latter order was more prevalent because, they thought, it was more representative of the population. But in fact, both orders are equally likely.

Or consider our partiality for “availability,” also known as “the endowment effect”—the preference we assign to familiar or prominent objects or ideas. Morey grappled with this bias when considering whether to trade backup point-guard Kyle Lowry to another team. As Lewis puts it, “the mere fact that they owned…Lowry appeared to have distorted their judgment about him.” Ultimately making the wildly successful trade, Morey instructed his lieutenants going forward to establish the draft-pick value even of existing Rockets players.

The researchers found a similar trend when asking respondents whether, in a sample of a random text, the letter “K” appeared more frequently in the first or the third position of any given word. Because most of those surveyed could more easily recall words beginning with “K,” they checked the first box. But in fact, “K” appeared in the third position almost twice as often.

To these tendencies, which Tversky and Kahneman labeled “heuristics,” can be added “anchoring,” “hindsight bias,” and “confirmation bias,” all of which the pair coined and/or inspired. But their most famous discovery, for which Kahneman won the Nobel Prize (Tversky surely would have shared the prize had he not died several years earlier), involved what they would later call “prospect theory”: a peculiar mismatch in human risk-aversion between situations of gain and loss that initiated their work in the field of behavioral economics.

They found, for instance, that people vastly preferred winning a certain $400 to a 50% chance of gaining $1,000—that is, they were willing to pay a $100 premium for the sure thing. But when faced with the choice between a certain loss of $400 or a 50% chance of losing $1,000, they overwhelmingly chose the gamble. Their conclusion: our utility-maximizing decisions are heavily context-dependent. “What constitutes a gain or a loss,” they wrote in their breakthrough paper’s early draft, “depends on the representation of the problem and on the context in which it arises.”

Lewis masterfully blends the personal story of the rise and fall of Kahneman and Tversky’s collaboration with a sharp-eyed analysis of its fruits. In one especially poignant vignette on the pair’s fascinating “conjunction theory,” which they developed amidst a fracture in their relationship, Lewis observes that “the public perception of their relationship was now a Venn diagram, two circles, with Danny wholly contained by Amos.”

In the end, we’ll never fully appreciate how much we don’t know. But this groundbreaking collaboration, depicted so beautifully in Lewis’s prose, helps give us some idea.