A new model for different speed and accuracy requirements in pointing tasks

Fitts' law is a tool for evaluating pointing devices, which has been accepted and applied widespread in the human computer interaction field. However, there are still some problems embarrassing the researchers in this field about its validity. One problem is derived from the request on input hits' distribution (i.e. spatial constraint) by the origin of Fitts' law. Therefore, a new model based on temporal distribution was developed to alter the traditional models. We carried out an experiment including five tasks with different requirement on speed and accuracy to test the effect of the new model. The new model and the traditional models are compared with both the mixed data and the individual task data of the experiment using AIC (Akaike's Information Criterion), a criterion for statistical model selection. All results show that the new model is better than the traditional ones in performance evaluation.