How We Decide

Every day you make hundreds of decisions. Most are so fast you barely notice them — which route to walk, what to order, whether to check your phone. A few take longer: which classes to take, how to handle a conflict, whether to trust a piece of information. And occasionally a decision carries real weight: a career choice, a relationship, a commitment that reshapes your future. What most people never study is the process itself. Not what to decide, but how decisions are made — the cognitive machinery behind every choice. This lesson opens that machinery and examines it carefully. Understanding how we decide is the foundation of deciding better.

Two Systems of Thought

Psychologist Daniel Kahneman — who won the Nobel Prize in Economics for this work — proposed that human thinking operates through two distinct modes, which he called System 1 and System 2. System 1 is fast, automatic, effortless, and largely unconscious. It pattern-matches against experience. When you see '2 + 2', the answer '4' arrives without deliberation. When you walk into a room and sense that something feels off, that sense is System 1. It is the brain running on heuristics — mental shortcuts built from a lifetime of learning. System 2 is slow, deliberate, effortful, and conscious. It handles tasks that require careful reasoning: working through a multi-step math problem, evaluating a complex argument, weighing competing values. System 2 can override System 1, but it is metabolically expensive — the brain prefers to avoid using it when System 1 seems adequate. The trouble is that System 1 is fast partly because it skips verification. It produces answers that feel right without checking whether they are right. Decision-making errors arise largely because System 1 fires confidently on a problem that actually requires System 2 — and System 2 either never engages, or endorses the System 1 answer without scrutiny.

The Anatomy of a Decision

A decision always has the same structural components, even when they are invisible. First, there is the decision problem: a situation in which action is required and multiple options exist. Without genuine alternatives, there is no decision — only an instruction. Second, there are the options: the candidate choices. Good decision-making begins by being explicit about what alternatives actually exist, because people systematically under-generate options. The default comparison is often 'do this thing versus do nothing,' ignoring a dozen other alternatives. Third, there are outcomes: what happens as a result of each option. Outcomes are almost always uncertain — they depend on how the world behaves, which you do not fully control. A decision can be excellent even if the outcome turns out badly; a decision can be poor even if the outcome happens to be fine. Separating decision quality from outcome quality is one of the most important skills in this module. Fourth, there are values: what you care about. Outcomes are not equally desirable. Knowing your values — and being honest about which values are actually driving a choice — is essential to coherent decision-making. Fifth, there is the choice itself. Given your options, your beliefs about outcomes, and your values, you select an action.

Process Quality vs. Outcome Quality

The most important distinction in decision science is between the quality of the process and the quality of the outcome. These are logically independent. Consider a physician who follows all best-practice guidelines in diagnosing a rare condition but — because the disease presents atypically — the patient is not helped. The process was sound; the outcome was bad. Now consider a student who guesses randomly on a multiple-choice exam and scores 100%. The outcome was good; the process was terrible. Resulting is the error of judging the quality of a decision by its outcome — something you did not fully control — rather than by the quality of the reasoning process at the time. Resulting is seductive because outcomes are visible and processes are not. Poker players are acutely aware of this: making a statistically correct call and losing the hand does not mean the call was wrong. This distinction matters enormously for learning. If you evaluate past decisions only by their outcomes, you will draw the wrong lessons: you will abandon sound reasoning that happened to produce a bad outcome, and you will repeat poor reasoning that happened to produce a good one. Evaluating the process — the information available at the time, the reasoning applied, the options considered — is how you actually improve.