Three models to account for discrimination times are examined : one being derived from information-theory, another being Grossman's Confusion Function and the third a form of Sequential Sampling model. Four experiments, using card-sorting tasks in which lines of different lengths had to be discriminated, wore conducted and the results compared with those of previous experiments. It was concluded that : 1.Experiments in this field may be divided hi to two types. Ono type (A) involves a series of judgments the difficulty of which differs randomly from one to another. The either type (B) involves a series of judgments all of the same degree of difficulty. Type B results in the present experiments were fitted moderately well by Crossman's Confusion Function, and somewhat less well by the Sequential Sampling model. They were not well fitted by the information-theory model. Type A results had previously been shown to be well fitted by an information-theory model when each individual discrimination was separated in time from others. It was found that when discriminations were made as a continuous series, the relationship between difficulty and discrimination time broke down
