記憶痕跡の變化:主としてその研究法に就いて
スポンサーリンク
概要
- 論文の詳細を見る
Most experiments on memory have chiefly been concerned with learning, so little has been studied solely on the problem of retention, especially on the change of memory traces. Among a few experimental works, which have relation more or less to this problem, the one made by F. Wulf from the standpoint of "Gestalttheorie" is worth noticing for its simplicity in method and for its obviousness in resultant data. The method applied to his experiment is based on the assumption that the features of deviations observed in the graphically reproduced figures must exactly correspond to those of changes in memory traces of the originally perceived figures. It may, therefore, be said that, as far as the problem is concerned with the change of memory traces, nothing can be ascertained by his method, unless this assumption is verified. The purpose of the present experiment was to settle whether the validity of this assumption, and consequently that of the method based on it, could be verified or not.<BR>The procedure of our experiment was as follows:<BR>A stimulus figure (see Fig. 1, p. 846.) was exposed to a group of subjects for a minute. After a week one half of them (Group I) were made to reproduce it graphically. Various figures thus drawn by Group I were enlarged and arranged on a sheet of paper as they are shown in Fig. 2 (p.846.). Out of these, again a week later, all subjects, including both Group I and the rest (Group II), were told to select three figures, which seemed to them most like the original.<BR>The results were as follows:<BR>In the case of Group I, there could be observed two different tendencies in selection the one to select figures more or less like the original and the other to select those like the reproduction drawn by each subject. And, therefore, the degree of mutual resemblance between three figures selected by a certain subject varied in accordance with the degree of resemblance between his drawing and the original. The three figures selected by any subject of this group, who had drawn his reproduction much unlike the original, were widely different from one another, one or two of them being like the original and the rest being like the reproduction; so that it was wondered if any of them were considered to look more or less like the original. But those selected by any subject of this group, who had drawn his reproduction like the original, had much in common with one another and also with the original. (see Table. I, p. 848.)<BR>The matter, however, was simple in respect to Group II. Almost all subjects of this group made their selections from some three or four figures, which were more similar to the original than the others. (see Table. II, p. 848.)<BR>These results are quite contrary to those which ought to be expected from the assumption before-mentioned. It may, then, be concluded that the features of deviations observed in the graphically reproduced figures do not exactly correspond to those of changes in memory traces of the original, and consequently, a method based on the assumption, which admits this correspondence as such, is not suitable for the purpose of studying changes of memory traces.