Which version/variation of Fitts' law? A critique of information-theory models.

Four different versions of Fitts' law have been in use over many years in the ergonomics, psychology, and HCI (Human-Computer Interaction) fields. There appears to have been no critical discussion of the validity of these different versions. The question is asked: Is one better or more valid than others? The aim of this article was to evaluate these commonly used versions and attempt to determine the correct/more valid model. It is concluded that the original form developed by P. M. Fitts (1954) and P. M. Fitts and J. R. Peterson (1964) is the most valid form based on information theory and that the theoretical foundation of the Shannon version suggested by I. S. MacKenzie (1989) is invalid. A version due to A. T. Welford (1968) also has a valid information-theory basis.