Learning Fitts’ law with imprecise regression

The Fitts' law is used in order to predict the time for pointing a target with respect to its size and to the distance to cover. However, this law only predict the average time for pointing the target with a given difficulty. In this paper, we propose to use a fuzzy regression approach, named imprecise regression, which allows us to predict the general tendency (time with respect to the difficulty of the task), together with the imprecision associated with this tendency. In the first time we will present the imprecise regression approach. Then, we will discuss and compare the results obtained with linear regression and with fuzzy imprecise from the data obtained with a mouse pointing experimentation.