Accuracy of MHP/GOMS predictions for the task of issuing recurrent commands

This short talk examines the accuracy of MHP/GOMS predictions for a task involving new and revised estimates of several MHP operators. The study consists of a cognitive modelling component to generate predicted task performance times and an experiment to collect actual task performance times. I examine prediction errors, the extent to which the predictions meet the accuracy goal (i.e., predict performance times to an accuracy of 80% or more), and the suitability of the MHP estimates used. I also compare the predictions errors from this study with those from two other studies.