The present experiment is designed to examine systematically interactions of time, space and velocity in estimation at the field of motion perception. In this paper the author will report only the results of time and space estimations. The procedure is as follows: The standard stimulus, a light patch, moves from right to left on the right side of the screen, in accordance with the conditions shown in Table 1. One second after the standard stimulus disappears, the reproduction stimulus, another light patch, appears on the left side of the screen, when the exposure time of the standard stimulus is estimated. Subjects are requested to push the key when perceived exposure time of the reproduction stimulus equals to that of the standard. Two light patches used for reproduction of the spatial distance are presented one second after the standard stimulus disappears. Subjects are requested to adjust the interval between two reproduction stimuli to the spatial distance of the standard stimulus. Experiments of time, space and velocity estimations are done on different days. For each estimation there are four experimental sessions corresponding to four standard values, each of which contains several conditions of standard values (see Table 2). The orders among three kinds of estimations, among four sessions in each estimation, and among conditions of several standard stimuli in each session are all at random for each subject. Each subject is assigned a single trial for each condition of standard stimulus. The types and numbers of the subjects are in Table 3. Results are as follows: 1. Constant short physical times, that is, 1.1sec and 2.0sec, are significantly overestimated in all groups, as spatial distance and velocity become smaller, but as for constant long physical times, that is 3.6sec and 6.5sec, there is not such tendency (see Fig. 1, Tables 4, 6 and 8). 2. Constant short physical spaces, that is, 2.0cm and 3.6cm, are significantly overestimated, as exposure time becomes shorter and velocity becomes larger. This tendency is stronger for younger children and it is not significant for U group. For constant long physical spaces, that is, 6.5cm and 11.7cm, there is not such tendency in all groups. (see Fig. 1, Tables 7 and 8). 3. In all groups there are few subjects who show such relations as tau- and kappa-effects between time and space estimations, and contingency coefficients (C) between two estimations are almost zero (see Table 9). 4. In both estimations, reproduced values are larger than standard values, especially in the case in which standard values are small. In space estimation this tendency is stronger, as subjects are younger (see Figs. 1 and 2).
The quantitative measurement on the phenomena of “semantic satiation” has been improved by an application of the semantic differential (SD) method. However, it has also been indicated that there still remain several problems as to how to apply the SD method in such cases. The purpose of this study was to investigate some methodological problems in the measurement of “change of meaning”. In Exp. 1, three questions were presented: (1) to examine in detail the midpoint of a SD scale; (2) to classify the types of “change of meaning”; and (3) to measure separately “the change of meaning” on each scale. The following results were obtained. The interaction among the repeated words, the scale of meaning, and the types of change of meaning proved significant. It was clearly shown that word repetition may lead to transf ormative change (T), in addition to decreases of meaning (S) and increases of meaning (G). It was further displayed that the effects of word repetition on meaning were determined by the interaction among “words”, “scales”, and “types of change”. Exp. 2 was designed to examine the effect meaningless-word repetition. The results were found consistent to the Johnson's findings, in that the increases of generation change were obtained. In Exp. 3, the bipolar continuum of an SD scale was examined. It proved that “change of meaning” could be rated more accurately on a bipolar scale than on a unipolar scale. Besides, a few comments were given on the bipolar continuum on the basis of the Green's study. Exp. 4 was designed to confirm the property of “semantic satiation” by an application of the verbal association method. The results were found to correspond to those obtained by the SD method. In discussion, three interpretations (Inhibition, Learning, Adaptation-level Interpretation) presented by Amster (1964) were examined. Finally, the writer suggested the possibility of an application of the transformational function (by Berlyne) of mediation process.
We defined here the process of meaningful learning as follows: identifying equivalence (in the widest meaning) between incoming information and stored information, by transforming the former. An experiment of successive paired-associate learning was undertaken to compare the learning performance between 2 conditions of preceding learning (List 1 & 2), under one of which Ss could learn meaningfully, though Ss of both conditions were required to learn the same material. In one condition, Ss could learn meaningfully, i.e., they could find equivalence between stored information (lexical meaning of words) and learning materials of List 3, by displacing components in the response term of the List 3 (using associations learned in List 1 & 2) and integrating them by connectives; in the other condition, Ss could not find equivalence by such transformations and had to learn material of List 3 in rote manner. (See Table 1) 32 undergraduate students served as Ss. Results: the effectiveness of meaningful learning (as against rote learning) was clearly shown. (See Table 7) Furthermore, two different alternative explanations of the results were eliminated by using auxiliary lists. (See Table 3) The first alternative is that integration of R terms of List 3 could be facilitated by the fact that specific combinations of NS learned in List 2 were used as components of List 3. This facilitation effect was proved to be negligible, however, by the performance of control gr., Ss of which learned List 3′ which had the same R terms as, but different S terms from, List 3, so that the pair could not be learned meaningfully. (See Table 8) The second alternative is that the effectiveness of meaningful learning could be explained in more traditional manner, that is, by facilitating effect of mediated association and positive transfer by the common element. However this aiternative was also rejected by the performance of another control gr., Ss of which learned List 3″ which had the same S terms and the same components (NS) of R terms but had different inappropriate connectives. (See also Table 8)