Heuristics

Heuristics are cognitive shortcuts or rules of thumb that people use to simplify decision making and problem solving in the face of complexity, uncertainty, or time pressure. They reduce cognitive load by focusing on readily available information and approximate patterns, producing quick judgments that are often good enough but not guaranteed to be optimal.

In psychology, heuristics help explain how people form judgments and make estimates, but they can also lead to systematic errors or biases. Common examples include the availability heuristic, which bases probability judgments on how easily examples come to mind; the representativeness heuristic, which assesses similarity to a stereotype; anchoring and adjustment, which relies on an initial value; and the affect or simulation heuristics, which rely on emotion or imagined outcomes.

In computer science and related fields, heuristics refer to problem-solving methods that produce satisfactory solutions efficiently when exact methods are impractical. They include greedy strategies, local search such as hill climbing, and heuristic functions used in search algorithms like A* to estimate remaining cost. Heuristics in algorithms trade completeness and optimality for speed and scalability, and they may be domain-specific or general-purpose.

Applications arise in artificial intelligence, operations research, routing and scheduling, and usability engineering (heuristic evaluation). While useful, heuristics are context-dependent and can introduce bias or error if relied on beyond their validity. They are often combined with formal methods or empirical testing to assess performance.

The term derives from the Greek heuriskein, meaning to find or discover.