Abstract
This investigation studies the changes involved in the perception of some simple geometrical figures when presented in different sizes for a fraction of a second.
Nine geometrical forms, a square, a triangle, a circle, a pentagon, a hexagon, two rectangles (horizontal and vertical) and two ellipses (horizontal and vertical), each in five different sizes, were selected as stimulus materials. They were drawn in black ink on white card-boards, and were exposel. one at a time in haphazard order for 9 milliseconds by means of fall-tachistoscope and the observer was asked to describe exactly what he saw.
The experiment was conducted in a dark room. The light was furnished by two 60 watt type C Mazda daylight lamps hung back of the observer. The distance between the stimulus and the observer's eyes was about 2, meters.
The results for four observers who took part in the experiment may be summarized as follows:
I. When a geometrical figure is presented for such a short interval as in our experiment, the observer may either see or not see the stimulus. Even when he sees it, his visual excerience may not always correspond with the stimulus itself.
A. He may see something, but this something he cannot identify. He is merely conscious of an object, a gray blur, a sort of halo against a white background. We shall call this the halo-phenomenon.
B. He may perceive a figure and be able to name it, but his perception is a little disfigured. We shall call this the transformed perception.
C. He may see the figure exactly as it is given (perfect perception).
No matter what forms are used as stimuli, the relative frequencies of these different experiences are influenced by the size of the figure. The averaged R. L. for all the stimulus figures as expressed in terms of the area of the figure is 73.17 sq. mm.. The limen for the halo-phenomenon is 117.29 sq. mm., for transformed perception, 207.45 sq. mm. and for perfect perception, 245.21 sq. mm.. It would seem, therefore, that these different experiences represent the developmental stages of perceptual processes.
Further, the values of the limen are different for different forms. For instance, the average R. L. for a triangle is about 1/2 of the R. L. for a hexagon, and the average of the perfect perceptual limen for a hexagon is 2.7 times as big as that for a triangle. (See Tab. III, IV, V, VIII, in the Japanese text, pp.439-448.)
II. Introspehtive reports.show that there are marked qualitative differences between the transformed and the perfect perception, the latter appearing in surface colours and theformer in film colours. The latter is distinguished by its compactness and its clear contours.
