Includes bibliographical references and index.

"This book is motivated by a fundamental puzzle about human cognition: how can we apparently be so stupid and so smart at the same time? On the one hand, the catalogue of human error is vast: we perceive things that aren't there and fail to perceive things right in front of us, we forget things that happened and remember things that didn't happen, we're inconsistent, biased, myopic, overly optimistic, and-despite this litany of imperfections-overconfident. In short, we appear to be as far as one can imagine from an ideal of rationality. On the other hand, there is an equally vast catalogue of findings in support of human rationality: we come close to optimal performance in domains ranging from motor control and sensory perception to prediction, communication, decision making, and logical reasoning. This puzzle has been around for as long as people have contemplated the nature of human intelligence, though it is now amplified by the modern revolution in AI. In this book, Samuel J. Gershman offers a new explanation, grounded in computational neuroscience, for this puzzle. He argues that the errors that the brain makes-those that make us "stupid"-are not haphazard "hacks" or "kluges" as some have argued. Rather, they are inevitable consequences of a brain optimized to operate under natural information processing constraints. In this book, Gershman develops this argument and shows how it reveals a deeper computational logic underlying a range of errors in human cognition. Importantly, he does not develop a bespoke explanation for each individual error; rather, he develops a uniform computational logic that can be invoked to explain diverse and superficially distinct phenomena. The result is a small set of unifying principles for understanding both the successes and the failures of cognition"-- Provided by publisher.

"How a computational framework can account for the successes and failures of human cognitionAt the heart of human intelligence rests a fundamental puzzle: How are we incredibly smart and stupid at the same time? No existing machine can match the power and flexibility of human perception, language, and reasoning. Yet, we routinely commit errors that reveal the failures of our thought processes. What Makes Us Smart makes sense of this paradox by arguing that our cognitive errors are not haphazard. Rather, they are the inevitable consequences of a brain optimized for efficient inference and decision making within the constraints of time, energy, and memory-in other words, data and resource limitations. Framing human intelligence in terms of these constraints, Samuel Gershman shows how a deeper computational logic underpins the "stupid" errors of human cognition.Embarking across psychology, neuroscience, computer science, linguistics, and economics, Gershman presents unifying principles that govern human intelligence. First, inductive bias: any system that makes inferences based on limited data must constrain its hypotheses in some way before observing data. Second, approximation bias: any system that makes inferences and decisions with limited resources must make approximations. Applying these principles to a range of computational errors made by humans, Gershman demonstrates that intelligent systems designed to meet these constraints yield characteristically human errors.Examining how humans make intelligent and maladaptive decisions, What Makes Us Smart delves into the successes and failures of cognition"-- Provided by publisher.

Introduction: are we smart? -- Rational illusions -- Structure and origins of inductive bias -- Learning from others -- Good questions -- How to never be wrong -- Seeing patterns -- Are we consistent? -- Celestial teapots and flying spaghetti monsters -- The frugal brain -- Language design -- The uses of randomness -- Conclusion: what makes us smart.

Description based on online resource; title from PDF title page (JSTOR, viewed on October 22, 2021).

IEEE IEEE Xplore Princeton University Press eBooks Library