It has been written among others by Volkmann, Gösser, .Frank, that negative after-images were influenced by the contour figures on the projection surface. But they have not studied systematically enough these phenomena as such. It is the purpose of this study to investigate these phenomena and, if could, to find out some explannations for them. At first, I had my subjects project a circular after-image on a homogeneous surface. Then the after-image grew fainter and fainter from edge to centre, and became at last a small circle and disappeared. But once projected on a straight line, the same circular after-image was, so to speak, absorbed into the contour line beginning to change the form at edge. Moreover when a closed contour figure was drawn on the projection surface, there was a phenomenal difference in the behavior of the after-image between inside and outside of the figure. Generally speaking, the contour figure influenced the after-image in the direction in which the figure phenomenally appeared. According to W. Köhler, the phenomenal appearance of the contour figure must be the psychological picture of a stationary equilibrium distribution in the corresponding processes of the brain. So, it seems to me that the after-image is influenced by a stationary equilibrium distribution in the processes of the brain. If this assumption is true, it is expected to occur that when one project the same after-image on the varied contour figures, there will appear phenomenal differences respectively. For the distributions corresponding to the varied contour figutes must be very different. In order to test this, I had my subjects project a circular after-image in turn on the centre of circle, square, and triangle of equal area. The results in each case, as might have been expected, greatly differed from each other. In the next place, I wanted to see the effect of the position of projection in the contour figure upon the after-image and had my subjects project a little circular after-image On the varied positions of a square (centre, near a corner, near a side). The results showed that the after-image moved and disappeared in very different ways according to the differences of the positions. The effect of “Figure” and “Ground” upon the after-image was tested also, and the results served to confirm the above-mentioned assumption. Now, we face here on a problem; why after-image is influenced by the distribution corresponding to the contour figure? I could ascertain in a preliminary series of experiments the following facts that the forces of distribution corresponding to the after-image were weakened more and more as time passed by. While, on the contrary, the forces of distribution corresponding to the contour figure do not get weakened as time passes by, and have always fairly constant strong distribution. Such being the case, when these two forces are brought about simultaneouslY at the pSychophysical field, they affect each other, and it is natural that the weaker force (after-image) should be influenced by the stronger one (contour figure).
Da die frühere Untersuchungen fiber die Grössenkonstanz von dein Punkte aus behandelt warden waxen, dass die Sehgrösse und das Retinalbild nicht immer korrespondierten, versuchte man die Sehgrösse phdnomenologisch unter den verschiedenen räumlichen Bedingungen zu messen, irgend einen Moment zu linden, der die Sehgrösse damit bestimmen kann. Abet was bei diesel Versuchung der Frage den Vpn. belastet werden soil, ist der Vergleich der beiden Objekte, so lies es sich urn die Vergleichsprozesse in der ganzheitlichen Raumstruktur, weld.e die zwei Objekte und Vp. einschliesst, handeln. Da wit nun zuerst das Experiment fiber den Zusammenhang der Sehgrösse und des phanomenalen Abstandes ausführten, ist es uns aufgestellt, dass kein eindeu-tiger, funktionaler Zusammenhang dazwischen vorhanden 1st. (Exp. 1) Esischiermilir, dass eernene bedeutende Rolle spielen wfirde, ob der Raum dabei bezüglich der Sehgrösse einheitlich aufgefasst werden konnte. Da ich diesnmum zu feststellen die Verhältnisse des Raums, in dern die Ciegenstände vorhanden sind, mit verschiedener Weise verändernd versucht habe, beobachtete ich, dass, wenn das Feld des Vergleichs phanomenal einheitlich ist, nämlich die Gegenstande dinghaftig gesehen, und die Zusammenhang zwischen ihnen festig behalten wird, und das Ganze gut gegliedert ist, dann vor allem die Grössenkonstanz sich gross befindet, sonst die Erscheinungsweise des Objektes labil ist und eine spezielle Veränderung über die Sehgrösse gefunden wild. Und ausserdem ich fand, dass diese Veränderung eine ähnliche Richtung mit der induzierten Bewegung hat. (Exp. 2-6a) Damit der Vergleich nun ausgefuhrt werden soll, muss irgend ein Zusammenhang zwischen den Objekten existieren, so nennen wir das die Verankerungsfläche, auf die das Vergleich geschlossen wird. Da unter der Aufgabe des Vergleichs, die vergleichene Dinge entweder phänomenal sowie möglich miteinander nahe zu liegen oder in dem gleichwertigen Abstand von Vp. zn sein, bessere Bedingung ist (Takagi, K.: Jap.J. psychol., XI, 1), so wird die Verankerungsfläche mit “ich” in frontalparallele Verliältnisse herankommen. Und we es wirklich ausgeführt werden kann, z. B. man strecke seine Arme aus, womit man mit der Länge der Finger die Länge des entfernten Objektes messe, in dem soll es gedacht werden, dass das entfernte Objekt zum Finger lokalisiert und die Grösse geometorisch-optisch bestimmt wird. Aber wenn die Konstanz gross ist, so ist die Lokalisation des Objektes im allgemeinen fest, deswegen soll “Ich” relativ an die Verankerungsfläche verschoben werden. (Köhler, W.: Psychol. Fcrsch. 18) Diese Verhältnisse sind schematisch in Fig. 4 (Siehe Japanische Text S. 256) gezeichnet. Dann wir haben nach dieser Schema die Verhltanisse der phänomenalen Festigkeit bezüglich der Stelle des Objektes (Exp. 7), und des Unterschieds der Festigkeit aus dem Gestaltzusammenhang des Objektes (Exp. 8) mit der Grössenkonstanz versucht und die Resultat der Annahme zufrieden war. Nun die Frage über die Grössenkonstanz scheint uns dann aufgeklärt zu werden, wenn die Dynamik der Vergleichsprozesse in dem dreidimensionalen Raume in die Klarheit gebracht werde.
Experimental studies of the “so-called” form constancy phenomenon were investigated by R. Thouless (Brit. J. Psychol., 21, 22) and K. Fissler (Arch.f. d. ges.Psychol., 88), and their explanations of the phenomenon were almost the same in that the perceived form was a compromise between the real form of the object and retinal-image of it. In the first place I started to examine the adequacy of these explanations (Exp. 1). The standard object (KO) was a white cardboard circle of 20cm. diameter, exposed with an inclination of 30 from the frontal-parallel plane, and variable objects (VO) were a white cardboard circle of 20cm. diameter and fourteen ellipses, (Their horizontal axes were always 20cm.), presented frontal-parallelly. They were presented in front of the frontal-parallel background of grey paper. From the result of this experiment their explanations seemed to be adequate. But at the same time we have an experience that, when we look at a tea-cup obliquely, we usually see it almost circular. From this kind of observation we can suppose that, if KO is presented singly, the perceived form of KO is different from that of Exp. 1. Making experiments on this assumption (Exp. 2), it was ,found that our assumption was right. Moreover there are some weakpoints in Eissler's study, for example; the problem of over-constancy. So we need a new wider assumption-and it seems to me that K. Koffka's assumption is exactly fitted for this purpose. He assumes that the form and the orientation are an in variant relation, i. e., constancy will be the greater, the more adequate the perceived orientation of figure is. I examined this assumption (Exp. 3), and gained the result which approximately verified it. It is a question, however, that the perceived orientation is always constant no matter whether both KO and VO are presented or not. Investigating this point (Exp. 4), I could find that it was different in these two cases. Hence we must divide our problem into two parts, i. e., constancy under comparison and that under single presentation. I took up the former as our problem at the present, and accordingly it seemed natural that on the comparison of KO and VO the phenomenal localization of VO influenced the perceived KO. Under this supposition, I introduced some changes in temporal and spatial distances between KO and VO (Exp. 5), and made the phenomenal localization o VO unstable (Exp. 6). From the results of these experiments I could verify our supposition. In fact, phenomenal localizations of KO and VO are determined by the whole stimulus-constellation, and if we change a part of it, phenomenal localizations will change also. Accordingly if we incline the background at the angle of 30°, 45° from the frontal-parallel plane (Exp. 7), we can expect the change of the degree of constancy and the results of experiments showed that our expectation was fulfilled. Then it was asked in what way that the process of comparison were conducted. As a means of finding a clue to it, I made some experiments of comparison with line-figures. Considering the results of them and W. Kohler's work (Psychol. Forsch., 18), it could be concluded that the comparison was accomplished in our case too by the “Verlagerung” of “Ich”. So we can conclude that the perceived form of KO is determined by the process of comparison between the phenomenal localizations of KO and VO, and the process of comparison itself determined in its turn by the dynamics of cerebral rocesses corresponding to “Ich” of O and his enviroment which contains the two objects.
Problem: The perception of one's own movement has been ascribed to the hunction of semicircular canals or muscular sensations by the physiological investigators, but it is possible to treat it from a different point of view, i. e., as relational phenomena between man and his Umfeld in a certain field. Experiment I : The subjects were instructed to judge the direction of their movements during actual movements in various degrees of speed and directions in a elevator. They were requested to shut their eyes while experiment was going on. Result: This condition of closed-eyes set forth a comparatively homogeneous optical Umfeld. And the given field for the bodily movement can be thought as somewhat vague one. Under these conditions the subject became very careful for his own body and his perception of movement seemed to be considerably determined by the degree of speed. Experiment II: The subjects were to judge the movement phenomena while looking at the outer wall through a trapezoidal prism, by means of which the direction of apparent movement of the wall was changed in various angles (0°, 45°, 90°, 135°, 180°) from the actual one. Result: In this case, the movement seemes to take place in a heterogeneous Umfeld. The subject was not very cautious for his own body, and his movement was mostly determined by the optical counter-direction of the movement of the outer wall. But when the accerelation was great, the course of apparent movement took the direction of the compound force of these optical and muscular sensations. Experiment III. A: After a certian period of movement (about 5 sec.) with his eyes; closed, the subject was to look out through the prism set in various angles. Result: When the change of the angle was great or the speed was high, the influence of accerelation was also great, while when the angle was small or the speed was low, the movement was almost entirely optically determined. Experiment III. B: About 10 minuites after the movement began which was optically given through the prism, the speed was altered and the subject was requested to close his eyes. Result: There appeared two cases. In the first case, the course took an intermediate direction between that of the accerelation and the course of the former movement. In the second, the course became intermediate only at the moment when the force was given and then it again returned to the former course. Experiment IV: About 10 minuites after the lateral movement through prism began, the subject, while sitting on a pivot-chair, was to point the directions of movements in various angles made by rotation of the chair with eyes closed. Result; When turned round by himself, there were two cases, namely the subject pointed either the direction adhering to the elevator or the direction adhering to himself. When turned round by others, there were also two cases, namely the subject pointed either a certain direction which was just a bit nearer to the direction of rotation from that of the elevator or always the direction adhering to himself. Conclusion: Under these experimental conditions the perception of one's own movement seems to consist in the reversion of one's own Umfeld. In a comparatively homogeneous optical Umfeld, the determination of its reversion is effected by the force given on his body, while in a heterogeneous optical Umfeld, the determination of the reversion, seems to be carried out optically.