ticular word anticipated, and with the reception of the word the release of the motor consequences (temporarily inhibited) of this "set," which already contains the movement in embryo.
Inner speech continued throughout experimentation, but in the fully developed "set" it cannot be regarded as the motor cue. That it cannot by itself be regarded as the cue is shown by the fact that not infrequently the movements outstrip the inner speech for the corresponding letters. That neither inner speech nor any other single factor can be regarded as in itself the cue is further demonstrated by frequent instances in which the movements were ahead of any sort of consciousness of the process. In fact, with both E and R in the last days of practice the moter impulse was at times clearly ahead of all detailed conscious processes. When consciousness of the movement came into the process at such times it always came in just a little behind what the fingers were doing. But this is not to be interpreted to mean that consciousness had nothing to do with the movements and that the latter had become purely physiological. Far from it. There was a distinct limit to both the separation in time and the nature of the possible withdrawal of consciousness from the movement. If in rapid writing the movements succeeded each other so rapidly that the consequent sensations merged into a sensation-mass and inner speech became indistinguishable or some of the letters were slurred or omitted, the "set was broken and control was momentarily lost. Again all the subjects found it possible at the close of experimentation to think of other things while writing, without interference. But here, too, there was a limit. It appears that, as one subject observed, considerable quasi-irrelevant imagery may enter consciousness without serious disturbance, provided it is all related to the writing. This irrelevant, but non-disturbing, imagery is only momentary "like a flash" or "like a dream." Otherwise the writing "set" is disturbed, if not actually broken up.
This brings us to another point in the discussion, namely the difference between the ' 'set' ' with eyes open and with eyes closed. There are two differences, a difference in the rate of the writing and a difference is distractability. Some of the subjects were inclined to think, on purely introspective grounds, that they wrote as rapidly with eyes closed as with eyes open. The facts (as revealed by the special records of the "practice" writing taken for this purpose without the subjects' knowledge) are, however, that in every case the eye-closed practice was slower than the eye-open; and this in spite of the fact that only two of the subjects visually regarded the card-board containing the letters. This is shown in the following table of averages
