ミュラー・リエル錯視外向圖形に於ける經驗の効果

抄録

The purpose of this study was to determine the effect of an interpolated task in decreasing the amount of the Müller-Lyer illusion.
The experiment consisted of three parts: In Part 1 the amount of illusion of one of the Müller-Lyer figure, the line with the arrow feathers (cf. Figure 1) was measured, in Part 2 some task given to the observer for 12-20 minutes and in Part 3 the amount of illusion of the initial figure was again measured. The experiment took place in the above order.
If statistically significant difference is to be found between the initial amount of illusion and that after a given interpolated task, it may be considered as the effect of the interpolated task. The effects of various kinds of interpolated tasks were examined in this way.
The principal interpolated tasks which were found to be effective in decreasing the amount of illusion were:
Task A……repeating the adjustment trials many times on the Müller-Lyer figure with the arrow feathers.
Task B……looking for the little dark spots presented one by one on different parts of the illusion figure (cf. Figure 3). The possibility of comparing the illusion figure with the variable line, during the task was excluded in later experiment.
Task K……drawing the illusion figure many times on a sheet of paper.
Task L……a task similar to A, but the illusion figure used here was one consisting of a line with the arrow heads.
Task N……a task similar to B, but the illusion figure used here was one similar to that used in L.
The interpolated tasks which were found to be ineffective were:
Task F……looking fixedly at about the middle part of the illusion figure.
Task I……a task similar to B, but the positions where dark spots were presented were limited to the intersecting points of the obliqus lines (cf. Figure 5).
In addition, the author tried to test Köhler-Fishback hypothesis by using a figure shown in Figure 8. It was found that mere prolonged fixation on a given fixation point (a little cross mark on the center of Figure 8) itself had no significant effect in decreasing the amount of illusion, whereas the observation shifting freely on every part of illusion figure was remarkably effective.
These results have led the author to the following conclusions:
1. Careful observation of every part of the illusion figure is as effective in decreasing the amount of illusion as the repeated ‘practice’ of adjustment trials. The author would call this effect “the effect of experience”.
2. In this experiment on “the effect of experience”, it is important to inspect carefully the whole figure structure, especially the part of oblique lines; the observation of the connecting line is not so important.
3. The figure used for purpose of giving “the experience of seeing that figure” needs not to be the same as the initial figure. It is necessary, however, that the two figures should have some structural character in common with each other.
4. Existence of “the pattern of satiation”, as mentioned in the Köhler-Fishback theory, might not be always the necessary condition nor always the satisfactory condition for bringing the effect stated above.
From these conclusions, the author would like to suggest the necessity of postulating such an intervening variable as determines the interaction between the ‘percept’ and the ‘set’, and at the same time, is determined by the experience in ‘some’ aspects of the ‘percept’.