Cognitive Scaffold

Preparing your thinking workspace

arrow_back_ios_new
MENTAL MODEL · M6644

Goals, Operators, Methods, Selection Rules

Goals, Operators, Methods, Selection Rules
Learn & MetacognitionHigh supportCognitive Science
Included
account_tree

Updated 2026-08-09

Loading revision record…

INTRODUCTION

English translation pending.

CORE DEFINITION

GOMS, standing for Goals, Operators, Methods, and Selection Rules, is a task analysis model introduced by Stuart Card, Thomas Moran, and Allen Newell in 1983. Goals are the states the user wants to reach. Operators are the elementary actions available, such as clicking, typing, or looking. Methods are the learned sequences of operators that accomplish a goal. Selection rules determine which method applies when more than one is available. Because the model represents a task as a formal structure of elementary actions, each with an assumed execution time, it can predict total task time and identify which steps dominate it, which makes it a tool for comparing interface designs before building them.

SCAFFOLDING EFFECT

psychology

Reduce cognitive load

- Use goal decomposition: state the user's goal, then break it into elementary actions. - Use method comparison: write the operator sequence for two designs and compare total time. - Use bottleneck hunting: find the step with the highest cost and target it for redesign.

anchor

Anchor fast decisions

Predicting performance requires a model of the task rather than a test of the product, and GOMS supplies one by reducing a goal to a hierarchy of operators with known durations. Summing those durations gives a time estimate that can be compared across designs without building either one. The estimate also localizes cost: whichever operator appears most often or takes longest identifies the step worth optimizing. The model's assumptions are its limits, since it counts motor and perceptual operators but treats the user's knowledge and method choice as fixed.

MINIMUM ACTION

In progress 0/1

Practice this model in one real situation:

Check to track your progress (stored locally)
Learning progress0%
account_treeGenealogyexpand_more
menu_bookReferencesexpand_more

Source support: Explicit

RELATED MODELS