抄録
Problem: In the fourth of a series of papers the writer has reported an investigation upon the general law of psycho-physical induction. In connection with this problem a further detailed analysis will be made in the present article.
Now, the outside distances of the diagram A, B, C, etc. in Fig. 7, although actually equal, appear to be of unequal length.
In the following investigation, I have sought to determine a relation between the amount of the illusion and the size of the inner interpol ated stimulus which varies from zero to a marked degree of length.
In making the measurements, the stimulus diagram was observed at the distance where the outer dots of the diagram are just visible. (The size of each dot is 0.5 mm. in breaatn and 1mm. in height. The observer usually sat at a aistance of 5 meters from the objects.)
Fig. 1.
Results: Fig. 2 in Japanese text p.454 shows a typical result of the experiments. It will be seen from the figure that the amount of the illusion varies quite regularly with the change in the length of the inner line. The over-estimation decreases with the increase in the size of the inner line, and the illusion is reversed when the inner line is sufficiently long. The change takes place slowly at first, then rapidly, and finally slowly again. This is what in mathematics is called a sigmoid curve.
There are many variable factors which govern the course of the illusion curve. In the first place, the curve shifts according to whether dots or lines are used -as the outer picture. Other variable factors are the size of the diagram, and the distance between the outer dots (or lines). Fig. 7 in Japanese text p. 460 shows how the illusion varies with the size of the diagram. The larger the diagram, the smaller the over-estimation. Finally, the decrease of over-estimation takes place as the distance between the outer dots is increased (See Fig. 4 in Japanese text p. 456)
To account for these phenomena a physiological hypothesis was offered. According to the writer's supposition, the psycho-physical induction has both peripheral and central causes, and it is quite possible that the synaptic connection of the nervous tracts plays an important rôle.
