Decision making shapes our lives, yet humans systematically depart from perfect rationality in remarkably consistent ways. We negotiate salaries, choose investments, hire employees, and make medical decisions daily, each choice shaped by cognitive limitations we rarely acknowledge. Understanding how our minds actually work, rather than how economic theory says they should work, offers a pathway to better choices and better design of the environments in which we choose.
The study of judgement and decision making reveals that our departures from rationality are not random errors but predictable patterns. Max Bazerman of Harvard Business School argues that recognising these patterns is not pessimistic; it is liberating. Once we understand the systematic ways our thinking misleads us, we can compensate, design around our limitations, and make deliberately better decisions when the stakes matter most.
Bounded Rationality and Cognitive Limits
Herbert Simon, in work that earned him the Nobel Prize, introduced the concept of bounded rationality to explain why humans cannot behave like perfectly rational economic actors. Simon observed that we genuinely try to be rational, but we face insurmountable obstacles. Our cognitive capacity is limited; time is scarce; information gathering is costly; and our memory is fallible. Rather than optimising every decision, we satisfice, meaning we search for solutions that are good enough rather than perfect. This is not a flaw in human thinking but a rational adaptation to the real constraints of human existence.
Understanding bounded rationality shifts how we interpret decision making. We are not irrational; we are rationally bounded. This distinction matters because it suggests that better decisions emerge not from scolding people for being illogical but from acknowledging real limits and working within them.
Systematic Biases and Mental Shortcuts
Amos Tversky and Daniel Kahneman, also Nobel laureates, moved beyond Simon to identify the specific, systematic biases that plague human judgement. They discovered that our minds rely on heuristics, mental shortcuts or rules of thumb that usually serve us well but systematically lead us astray in predictable ways. A heuristic allows us to make quick decisions with incomplete information; most of the time this efficiency saves us. Yet in certain contexts, these same shortcuts produce reliable, reproducible errors that affect our choices in ways we do not notice.
The study of biases and heuristics transformed psychology and economics. Where Simon explained that we are limited, Tversky and Kahneman explained exactly how that limitation manifests in our actual thinking patterns.
The Rational Decision Process
Bazerman and Moore articulated a six-step process that represents fully rational decision making. First, define the problem clearly. Second, identify the criteria that matter. Third, weight those criteria according to their importance. Fourth, generate a full set of alternatives. Fifth, rate each alternative against each criterion. Sixth, compute the optimal choice by comparing weighted scores. This process is systematic, transparent, and defensible. Yet most people rarely follow it, especially under time pressure or cognitive load. Understanding this ideal process matters because it reveals where our actual thinking falls short.
Three Classic Demonstrations of Bias
Overconfidence emerges when people are asked to set ranges within which they are ninety-eight percent confident a true value lies. Remarkably, their ranges capture far fewer than ninety-eight percent of actual values. People systematically overestimate the accuracy of their knowledge, leading to excessive risk taking and poorly calibrated predictions.
Anchoring occurs when an initial number, even one that is arbitrary or random, exerts a powerful pull on our final estimate. We adjust away from the anchor insufficiently, leaving our judgment biased toward the starting point. In salary negotiations, the first number offered often determines the final outcome not because it contains valid information but simply because it serves as an anchor.
Framing demonstrates that identical choices presented differently produce opposite preferences. Tversky and Kahneman presented the avian disease problem to participants: a disease threatens six hundred people; two programmes are available. When programmes were framed as lives saved, people chose the certain option. When identical programmes were framed as lives lost, people chose the gamble. We display risk aversion for gains and risk seeking for losses, a finding that contradicts rational choice theory and reveals how presentation shapes decision making.
Bounded Minds Beyond Rationality
Thaler and Bazerman with colleagues identified further ways our decision making remains bounded beyond the original rationality limits Simon described. Bounded willpower means we overweight the present relative to the future, which explains why we fail to save adequately for retirement despite knowing better. Bounded self interest reveals that we care about others outcomes, not merely our own gain, shaping choices in ways pure self interest cannot predict. Bounded ethicality describes how our ethical decision making is limited in ways we fail to recognise; we compromise our own values without noticing, a form of self deception. Bounded awareness means we fail to see obvious, important information that is available to us, a blindness to data that should influence our choices.
These bounds suggest that human limitation extends across domains: rationality, willpower, concern for others, ethics, and awareness itself.
System One and System Two Thinking
Stanovich and West distinguished two cognitive systems. System One operates fast, automatically, intuitively, and emotionally; it is the mind working without conscious effort. System Two is slow, conscious, effortful, and logical; it is deliberate thinking. The six step rational process belongs to System Two. Yet busy people, rushed and cognitively loaded, rely on System One, where heuristics and biases flourish unchecked. The practical implication is clear: shift to System Two for important decisions. Create conditions where System Two can operate, removing time pressure, reducing cognitive load, and making the decision process explicit.
Real World Applications and Institutional Design
Michael Lewis chronicles how Billy Beane of the Oakland Athletics replaced biased intuition with data analysis in Moneyball, demonstrating that systematic thinking outperforms gut feeling even in domains where experts claimed intuition reigned. Thaler and Sunstein explored how decision architects reshape choice through Nudge, altering defaults and environment. Automatic enrolment in retirement savings raises participation dramatically compared with requiring active enrolment; opt out organ donation produces substantially higher consent rates than opt in systems, as Johnson and Goldstein demonstrated. These examples show that fixing bad decisions does not always require changing people; it often requires changing the structure of choice.