so-called antagonistic reaction is intimately involved in these questions special study was given to it. The importance of the study of forms of reaction movements as distinguished from the mere investigation of reaction times and their introspective accompaniments was first brought prominently to light by Smith (5) in 1903. He had noticed in earlier reaction work that a reactor, in responding to the stimulus, does not always lift his finger immediately from the key on hearing the stimulus, but often presses down before making the final upward movement. This type of reaction Smith called "antagonistic reaction movement." In order to obtain graphic records of such a movement, Smith used a piece of stiff rubber tubing. Making one end air tight he connected the other with a recording tambour by a rubber tube. The reactor placed his finger on the stiff tubing and reacted by lifting the finger. The form of the reaction movement was thus recorded on the smoked paper of a kymograph. By means of a time line from a 100 v.d. tuning fork the time between the giving of the stimulus and the beginning of any movement, as well as that between the giving of the stimulus and the beginning of the final upward movement could be determined, the difference between them showing the time added to the reaction by the antagonistic movement. Smith reports the results from 33 subjects, 25 men and 8 women. Of these subjects he found 5 fairly constant antagonistic reactors, while 5 were intermittent in this type of reaction. Eighteen of the subjects gave no sign of the antagonistic movement. The remaining 5 were difficult of interpretation. The average reaction time, i.e., to the beginning of the final upward movement, for sensory reactions for 3 antagonistic subjects was 2000, 170o, and 180o, respectively. Corrected to the beginning of movement, these same subjects gave 180o, 1400, and 1400. The antagonistic portions of the reactions were, therefore, 200, 300, and 40°. One result of this investigation was, certainly, to show the obvious value of the study of the form of movement in reaction experiments, and this can be done only by the graphic method of recording. The importance of this general result is emphasized by a further fact, namely, that the form of reaction may change during the course of an experiment, for Smith found that the antagonistic reaction does not always appear at the beginning of the series. It may not appear until several series have been taken. Smith then carried on some extended experiments in order to determine the cause of the antagonistic reaction. He tried first the sensorial-muscular distinction in attention, He found that the antagonistic movement occurred in both forms of reaction, but that it was more frequent in the motor type. Experiments on the influence of distraction and fatigue on the antagonistic reaction gave no univocal results. Reactions were also taken with different sets of muscles, with different positions of the arm, and with different fingers. The antagonistic reaction movement remained under all conditions. Smith says (p. 57): "In every direction of movement, up or down, to the right or left, there were observed preliminary movements in the direction opposite to that intended by the reagent. The movements are not equal in both cases, but the curves of both persons experimented upon show the phenomenon unmistakably." It was not due, then, to any special set of muscles or to any position of those muscles. Smith presents an explanation for the antagonistic movement both from the physiological and from the psychophysical standpoints. Physiologically, he bases his explanation on the now well known work of Sherrington, in which it is shown "that when one of the pair of antagonistic muscles is innervated, there is a simultaneous inhibition of the muscles opposed to it." With this cue, Smith reasons that the muscles already innervated would tend to persist in their activity after the stimulus and before the subsequent inhibition occurs. From the psychophysical standpoint he states that in the period preceding the occurrence of the stimulus the most prominent idea in the mind of the reactor is that of holding the finger down on the key. The stimulus comes as a kind of shock. This shock tends, he thinks, to emphasize the idea already in consciousness in the instant preceding the voluntary decision to lift the finger. He mentions cer tain cases of patients suffering from general paralysis in which he has noticed this preliminary tightening of the muscles before the final action. Smith suggests a number of problems for future work, largely along the line of a more definite training of the subjects under different experimental conditions, particularly those of attention. Other references to Smith's antagonistic reaction which I have been able to find outside the work on reaction movements done in the Yale laboratory are by Titchener (6, 352-355) and by Franz (2). Titchener, reviewing Smith's work and that of the Yale laboratory, suggests further problems which need to be worked out in connection with the antagonistic movement, e.g., the effects of practice, of definite instruction, of introspective control. He favors a priori the view that the antagonistic form of reaction will appear more often in the motor reactions, and states that he has tried some experiments with the press reaction, but that no antagonistic movement was found. The experiments were, however, so few in number that no general conclusions can be drawn from them. Franz found that the reaction times of a depressed or retarded patient were longer than the normal and, much to his surprise, that simple reaction times increased without any change in choice reaction times. This change was so marked that in certain series the choice reactions were shorter than the simple. Since there was no evidence that the subject had changed the character of his attention in the simple reactions, or that fatigue played any part, Franz believed that he might be dealing with the antagonistic reaction movement. Smith had already stated that he had noted preliminary contractions in partially paralytic patients. Franz made no attempt in his investigation to determine this point. In 1905 Judd, McAllister and Steele (3) published an investigation on the character of reaction movements. Graphic records of both simple and discrimination reactions were taken from 51 subjects a total of 964 records of simple reactions and 523 of discrimination reactions. A special apparatus was constructed, consisting of a reaction key attached to one end of a metal spring in such a way that when the spring was depressed the reaction circuit was made; when the spring was released the circuit was broken (pp. 143, 144). A lever attached to the spring near the key recorded every change in pressure directly on a long strip of smoked paper. Times were taken with a Hipp chronoscope. These authors were interested not only in the nature of the reaction movement subsequent to the giving of the stimulus and terminating in a lifting of the finger, but also in the movements made in the period between the warning signal and the stimulus. They show drawings of many forms of movement, all more or less complicated, which lead to the conclusion that there is "no such thing as a simple reaction movement." All these records show, too, complex forms of adjustment in the period before the stimulus which determine the character of the final movement. For purposes of analysis the total reaction period was divided into that before the warning, that immediately following the warning, that just before the stimulus, and the final movement. An analysis of all the records yields three main types of reaction (pp. 166, 167): I. "The balanced type," a balancing of the antagonistic muscles shown by waves in the record. 2. "The antagonistic type," in which the pressure gradually increases in the period before the stimulus or suddenly increases directly after the stimulus. 3. "The partial reaction type," in which the tendency to react overpowers the antagonistic tendency, resulting in a partial reaction before the stimulus. The authors believe, therefore, that their results support the theory of reaction types, if by types is meant "individual tendencies even in untrained subjects to adopt a particular mode of reaction" (p. 179). The form of reaction was found to change, however, with a change in the "character of the task required" (p. 173) and it was clear that introspective evidence of the character of movement was not to be relied on (p. 180). Concerning the relation between form of movement and reaction time (pp. 169 ff.), it was found that a gradual antagonistic pressure in the period preceding the stimulus was advantageous for quick reaction. The sudden antagonistic reactions were, however, cases of excessive effort which, properly applied, meant speed but often, improperly applied, interfered with speed. The sudden antagonistic movement did not, therefore, as regularly promote speed as did the gradual. So far as the relation between form of movement and consciousness is concerned, the gradual increase in pressure "would indicate a growing emphasis in consciousness of the necessity of holding the hand down until the stimulus arrived" (p. 181), while the sudden antagonistic movement at the end seemed to be a matter of nerve mechanism rather than the result of a conscious process (p. 182). Finally, as problems for future work, are suggested the training of subjects in a “particular mode of reaction" and the study of the development of habits of movement (p. 177). II. APPARATUS AND METHOD The present investigation takes up the study of the form of reaction movements where the authors last mentioned left it and has, in general, a fourfold purpose: following their suggestions, five untrained subjects were, without special instructions, carried through a long series of reactions to see, first, if they would develop definite forms of movement; second, if during such a practice series a change in the form of movement would accompany the reduction of the reaction times and mean variations. The results of two of these reactors, reported in the original thesis, are omitted for lack of space. They agree with the results of the three reactors here reported. Five other subjects were trained to keep their attention on the stimulus or on the movement to see, third, if any differences in the form of the reaction movement accompanied the differences in the direction of the attention. The results of two of these subjects are omitted because of the small number of reactions that could be obtained from them. In their few series there appeared no differences in the form of reaction movement correlative to the differences in the direction of the attention. work of six subjects is reported in this article. In all, then, the |