Textbook of Usability

CPM-GOMS (Cognitive-Perceptual-Motor GOMS, also called Critical Path Method GOMS) models the parallel operation of the perceptual, cognitive, and motor subsystems of the Model Human Processor. Unlike the KLM, which sums operator times sequentially, CPM-GOMS recognises that expert users overlap these activities: while the eyes scan the next target, the cognitive system processes previous input, and the motor system executes the previously decided-upon action.

CPM-GOMS represents tasks as PERT charts (Program Evaluation and Review Technique), with operations as nodes and dependencies as edges. The task completion time is determined by the critical path — the longest chain of dependent activities.

The landmark validation was Project Ernestine (Gray, John, and Atwood, 1993), which used CPM-GOMS to predict telephone operator performance with a new workstation at NYNEX. The model predicted the new workstation would actually be slower than the old one by 4% — a counterintuitive finding confirmed by field measurement.

CPM-GOMS is more accurate than KLM for well-practised expert tasks but requires substantially more analysis effort. It is typically used when:

• The task is highly repeated, so even small time differences matter
• Expert performance is the primary concern
• Quantitative predictions must be defended rigorously

Predictions are typically within 10–20% of observed expert times.