JAPANESE STUDIES ON THE SO-CALLED GEOMETRICAL-OPTICAL ILLUSIONS

Abstract

The so-called geometrical-optical illusion (geometrisch-optische Tauschung) was one of the most important problems of psychology during the second half of the last century. However , only a few of European and American psychologists today are seemingly studying it, while in Japan it has continuously interested psychologists for these three decades. As Boring (1942) said, "the problem of the geometrical illusions is not a special problem of space perception ." They will serve as useful cues to discover the general laws. In the present paper, the important experimental works in Japan which have been done quantitatively through psycho-physical methods will be mainly reviewed.** Theoretical works will be referred to only briefly. I. ILLUSIONS OF LENGTH AND DISTANCE Muller-Lyer Illusion By far the most famous illusion of all is the one discovered by Muller-Lyer in 1889. As shown in Fig. 1 A, B, a straight line, to the both ends of which inwards directing oblique lines are attached, is generally underestimated in length, and another straight line with outward oblique lines is overestimated. Heymans (1896) and Lewis (1909) found that the smaller the angle of obliques was, the larger was the illusion, when the length of obliques was kept constant; and that, as the length of oblique increased, the illusion increased to reach the maximum and decreased afterwards, when the angle was kept constant. The former fact was confirmed by Kido (1927) and Naka-gawa (1958) in Japan, except that Nakagawa found some decrease in the smallest angle. Regarding the second fact Hayami & Miya (1937) found the maximal underestimation in the inward figure with the obliques which were as large as a third of the distance to be judged , and the maximal overestimation in the outward figure with obliques whose length was a half of the distance (Fig.