Two experiments are designed to examine the mentally retarded children's grasp of relationship of figures-series from the mental developmental viewpoint, and to testify the meaning and nature of M A measured by Binet-Intelligence Scale as a developmental index of the retarded.
The purpose of Experiment I is to test the relation between intellectual development of normal children whose CA ranges are from 4 years old to 6, and mental development of the retarded whose CA ranges are from 6 to 14 and MA ranges are from 3 to 7 (Table 1).
The problems presented to the subjects are to arrange geometrical figures-series cards such as circle, triangle, square and picturesseries cards as tortoise, butterfly, goldfish in the fixed row according to the series rule (Fig. 1). The purpose of Experiment II is to certify the difference of method of presentation of cards and models or tortoise, butterfly, goldfish to the retarded children in various years of age, but in the same MA.
The main results to two experiments are as follows:
(1) The abilities of grasp relationships of figuresseries in the mentally retarded develop similar to normal children according to their Mental Ages, and especially in the developmental stage MA 5, the abilities increase rapidly.(Fig. 2).(2) But the retarded children's grasp of relationships of geometric figures-series and of pictures-series or modelsseries are profoundly differ from that of normal children. And when the cards or models are presented by turns, grasp of relationships in the retarded children brings about better results than in normal children.(3) By comparison the younger retarded whose average CA is 5 years old with the older retarded whose average CA is 12, the older get better results than the younger, though the MA of both groups are the same.
From the above findings, we conclude that the abilities to grasp relationships of series increase proportionally to MA, and these, therefore, are a developmental index of intelligence, but that the meaning and nature of the.MA of the retarded are much differ from that of the normal.

View full abstract

[in Japanese]

View full abstract

From a model of Goldberg's Item Ambiguity. Index, it is assumed that there is a strong relationship between the Ambiguity Index values and the item difficulty. The respcnse latency and the ratings of judgemental confidence have been used as the measure of the item difficulty. And the intra-individual variability in understanding the meaning of the item, related to the item difficulty, has been investigated.
Then, it is assumed that there are such relationships between these measures and the Ambiguity Index values as;
1) the items of high Ambiguity Index values will have longer response latencies,
2) the items of high Ambiguity Index values should be associated with lowered confiderice by the individual in his answers,
3) the items of high Ambiguity Index values will have higher intra-individual variability in understanding the meaning of the items.
Fifteen items of high Ambiguity Index values and fifteen items of low Ambiguity Index values have been taken from the results of the previous research. Subjects were 45 college students. In these items, the response latency, the ratings of judgemental confidence and the understanding of the meaning of items have been investigated twice at two weeks' interval.
The above relationships have been ascertained.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

The present experiment was designed to test the hypothesis that implicit reinforcement phenomena in classrooms are found under competitive situations but not under noncompetitive situations.
The Ss were 5th-grade children of four classes, consisting of 20 boys and 20 girls in each class. Learning materials were digit symbol problems which were made of eight symbols and eigh t digits (one through eight), and the Ss were required to write the digits corresponding to the respective symbols as quickly as possible. The Ss in two classes were given the task for four minutes a day for two successive days under competitive instructions, and those in the other two classes under noncompetitive instructions. The Ss in each class were divided into two subgroups based on the first day's performances and the number of both sexes. At the start of the second day, half Ss in two classes, one under competitive and the other nuder noncompetitive situations, were positively reinforced (praised) for their first day's performances and those in the other classes were negatively reinforced (reproved), while the the remaining Ss in all classes were not given any information about their performances. The latter Ss were assumed to be implicitly reinforced, either negatively or positively, by observing their classmates subjected to explicit reinforcements. Thus the four groups were constructed under each situation, respectively: the group that was explicitly subjected to positive reinforcement (EP group), one explicitly subjected to negative reinforcement (EN group), one implicitly reinforced by observing their negatively reinforced classmates (IP group), and one implicitly reinforced by observing their positively reinforced classmates (INgroup).
The performance increment on the second day (number of correct responses en the second day minus that on the first day) was used as a measure to test the effect of experimental treatments. Table 1 presents means and SDs of the performance increment for eight groups, and based on the data a three-way analysis of variance was conducte d, which consisted of competitive or noncompetitive situation (C-NC), positive or negative reinforcement (P-N), and explicit or implicit reinforcement (EI). The result of the analysis is shown in Table 2. A significant triple interaction reflects the fact that IP group performed better than IN group under competitive situation, while under noncompet itive situation no difference was found between IP and IN groups and EP group performed better than EN group. Thus the present hypothesis was confirmed. A tentative explanation was presented that under competitive situation Ss were more sensitive to the performance of their classmates than those under noncompetitive situation, and therefore, they were easily subjected to implicit reinforcements by observing explicitly reinforced classmates. An additional piece of information was obtained that explicit positive reinforcement was more effective than negative one for girls, but no differential effect was found between them for boys.

View full abstract