Heuristics-Model Thinking

Heuristics are explicit shortcuts that reduce search, information use, or computation. Some perform well in particular environments, including cases where estimating many parameters from limited noisy data hurts prediction. Their usefulness depends on the match between the rule and the task.

A heuristic is a specified shortcut for selecting or inferring something with limited search or computation. To evaluate one, state what information it inspects, when it stops, and how it chooses. 'Trust intuition' is too vague to test. A rule that ignores unreliable cues can avoid fitting noise, but it can also miss a decisive exception.

Compare the shortcut with a realistic alternative on the same task. Accuracy, time, information cost, and consequential mistakes all matter. The appropriate question is which rule fits this environment, not whether simplicity or complexity wins in general. Optimization can incorporate uncertainty, and noisy environments still contain information worth using.

When to use it

When repeated decisions require a manageable procedure and a shortcut can be tested against a clear outcome and realistic alternative.

How it can help

Specify a candidate rule and compare its accuracy, effort, and important errors with feasible alternatives on representative cases. Retain relevant information and define exceptions instead of assuming less information is always better.