Abstract
Even where time, Ss, and money are available for an extensive experiment on a valid experimental task, the estimation of a system value for highly experienced operators is not foolproof. The Ss must be run on the task until their performance shows little or no further improvement with additional practice. Asymptotes must be estimated for each individual S via cautious application of curve-fitting techniques, and these estimates averaged for the appropriate sample of Ss.
Get full access to this article
View all access options for this article.
