heuristic

A heuristic is a practical method for problem solving, learning, and discovery that is not guaranteed to be optimal or perfect but is sufficient for reaching an immediate goal. The term comes from the Greek heuriskein, meaning “to find.” In everyday use, a heuristic is a rule of thumb, an educated guess, or a shortcut that reduces complexity and speeds up decision making.

In cognitive psychology, heuristics are mental shortcuts that ease cognitive load in judgment and decision making. They help people make quick assessments but can lead to systematic errors or biases. Common examples include the availability heuristic (relying on information that is most readily recalled), the representativeness heuristic (judging likelihood by similarity to a prototype), and anchoring (relying on an initial reference point).

In computer science and artificial intelligence, heuristics guide search and problem solving when exact methods are impractical. They aim to produce good-enough solutions quickly rather than guaranteed optimal ones. Examples include greedy algorithms, heuristic evaluation functions in game playing, and domain-specific rules that prune search spaces. In AI, search algorithms such as A* use admissible heuristics to estimate remaining cost and improve efficiency, while local search and metaheuristics are used for complex optimization problems.

A heuristic differs from a strict algorithm in that it does not guarantee a correct or optimal solution; its effectiveness depends on the problem domain and assumptions. Heuristics are widely used across fields such as decision making, design, planning, and optimization, offering speed and scalability at the cost of potential inaccuracy or bias.