the letters on the cardboard," and is already visualizing the letters infrequently and remembers their places in terms of motor activity. Here we have first the elimination of direct vision of the letters and a little later the elimination of the visual memory images of the letters, at least to a degree.
R fell into the other method, that of learning the list of words early in the practice and thus freeing the eyes for constant use upon the key-board. This eliminated eye movement from list to key-board and vice versa. On the second day R speaks of a greater ease in execution in consequence of having now "a general idea of where to look for the letters." After a few sittings R still further eliminated direct eye-movements from letter to bulb, so that the pressing of the bulb followed immediately upon the fixation of the letter. R also reduced eye-movements by gradually ceasing to fixate each letter as the corresponding bulb was pressed, and instead fixated the central part of the cardboard with only slight fluctuations on either side.
D never made direct eye-movements from the letter on the cardboard to the corresponding bulb. His method was essentially the same as that of R, with the exception that he used the list of words for a longer time. He did, however, early begin to eliminate eye-movements over the cardboard and the fixation of each letter. At the second sitting he thinks there is some motor memory present and finds that when the movement is once started the "visual element becomes peripheral and almost drops out. "At his eighth sitting he "fixates only the central part of the cardboard and makes use of the outlying letters in peripheral vision." A day or two later he reports that he "did not have to look at the letters on the cardboard." This was on the day when eye-closed "practice' ' was begun. He was, however, still looking at the list of words, which practice continued for a considerable time after the words had been committed to memory.
These details have been given to illustrate what is meant by what has been called the "short circuiting" process. They illustrate the general principle of initial excess of activity which characterizes all learning processes, a principle to which voluntary movement offers no exception; for voluntary movement is genetically a learning process, though it differs from certain learning processes in this respect, that much of what is eliminated is not necessarily faulty but at the beginning useful and even necessary. It is therefore not so much an elimination incident to "trial and error" as an elimination due to what may be called a knitting up of the associative links between the various sensory and perceptual processes involved in the control.
