同心円錯視に及ぼす図形明度の影響に関する研究

Ever since Benussi studied, in 1902, the influence of figural lightness upon the Zöllner's illusion, no systematic study has yet been carried out on the figural lightness in the illusion of concentric circles. To study the influence of the figural lightness upon the amount of illusion, various combinations of achromatic figures were used in the present study. They consisted of concentric circles with different grades of lightness drawn upon backgrounds which were dark gray (Japan Color Research Institute Lightness No. 13), neutral gray (J. C. R. I. Lightness No. 15) and light gray (J. C. R. I. Lightness No. 17) respectively, These grades of lightness (in parentheses) were equivalent to Munsell Renotation N 3.57, N 4.69 and N 6.30.<br>1) On the neutral gray background the amount of the assimilation of the inner circle first rose up with the increase in the radius of the outer circle, but it decreased afterwards, showing a convex curve. The highest values of assimilation were obtained when the radii of the outer and inner circles were in the ratio of three to two, as observed in eight different combinations of lightness. These results coincide with those of the former experiments in which stimulus figures on white or black backgrounds were used. It was also found that the figural lightness was more effective upon the amount of assimilation than upon the shape of assimilation curves or the position of peaks in these curves (Fig. 1, Table 2).<br>2) Two main factors could be assumed to influence the amount of assimilation; one, the difference in lightness of the figure and background, and the other, the interrelation in lightness of the standard (inner) and conditioned (outer) circles.<br>As to the first factor, Benussi's formula (1) could partly be applied to the experimental results. As to the second, the lightness of the standard (inner) circle was more effective than that of the conditioned (outer) circle.<br>The maximum amount of assimilation was obtained by the combination of an inner circle of J. C. R. I. Lightness No. 11 (N 2.38) and an outer circle of No. 19 (N 8.22), and the minimum was obtained by the reverse combination (Tables 2, 4). It was assumed from these results that a white figure on a neutral gray background, would be more effective and less influenced, than a black one on the same background.<br>3) Either the above two factors were also effective if other backgrounds such as No. 13 (dark gray, N 3.57) and No. 17 (light gray, N 6.30) were used: the absolute lightness of the figures was a more effective factor independent of the lightness of the background.<br>It was also observed that perceptual structures of figures had an effective influence on the amount of assimilation (Fig. 3).<br>4) In the case of three concentric circles, the configurational factor was more dominant; i.e., the figures with an equal lightness tended to form a group so that they played a main part in affecting the amount of illusion (Table 6).