Previous studies have shown an induction phenomenon in the rat, i.e. a considerable decrement in behavior variability and facilitation of a running response, when reinforcement was presented after or between non-reinforcement trials in the semi-circular maze. The purpose of the present experiment was to determine whether or not a similar effect was found when the reinforced test trials were given after simple maze exploration instead of usual procedure of non-reinforcement. Thirty rats deprived from foods were devided into three groups (Group 2-days, Group 7-days, and Group Pseudo-extinction). After initial exploration and pretraining, Ss were given 20 exploratory trials per day in the semi-circular maze. This procedure was repeated for 2 days in Group 2-days, and for 7 days in both Group 7-days and Group Pseudo-extinction. On the following day, Ss in both Group 2-days and Group 7-days were given 20 reinforced test trials, and Ss in Group Pseudo-extinction received these test trials after 60 non-reinforced (exploratory) trials. The results obtained during the test trials were as follows: 1) When the data of Group 2-days and Group 7-days were compared with those of the previous study in which the same number of trials as Group 2-days and Group 7-days were given with reinforcement before the test session respectively, significant decrement of behavior variability was found in both groups. However, there was no significant difference in the running time between each of the compared groups. 2) When compared with the previous study, almost the same amount of induction-like phenomenon was found in Group Pseudo-extinction. These pseudo-induction phenomenon increased with the number of exploration trials. This tendency would be similar to the relationship found between the induction and the number of non-reinforcement in the previous studies. From the above results, it is suggested that induction seems to be closely related to the exploratory experience or recognition of non-reward but not to the internal factors, such as drives or frustration.
The present study was performed to examine some factors which affect the reaction time (RT). They included the following four conditions: 1. the low level of anxiety induced by an application of slightly unpleasant electric shock (ES) (Exp. II), 2. the high level of anxiety induced by a very unpleasant ES (Exp. III), 3. the backward conditioning (Exp. IV), and 4. the anxiety induced by the presentation of false-instruction without ES (Exp. V). Exp. I constituted the control in the present experiment. Each experiment included three periods except for Exp. V. In the first period, three light stimuli (red, blue, and yellow) were presented randomly as a RT signal. Ss were instructed to respond to each signal as rapidly as possible. RT was measured by the pressing of a key paired in position with each signal. A verbal sign “Ready?” was given as a ready signal several seconds before the presentation of each signal. This period can be regarded as a neutral situation. The second period: the red signal was reinforced by pairing with the unavoidable ES. ES was given immediately after the response to the red signal in Exp. II and Exp. III, but was given immediately before the presentation of the red signal in Exp. IV. No ES was given to any presentation of RT signals in Exp. I. This period (except in the case of Exp. I) can be regarded as an anxiety inducing situation. The third period: RT was measured in the same way as the first period. Here, the specific anxiety effect caused by the conditioning should distinguish itself from the general effect of the anxiety inducing situation in the second period, for the differentiated stimuli (blue and yellow) could act as neutral stimuli, whereas the reinforcement stimulus (red) should act as an anxiety inducing stimulus. Exp. V consisted of two periods which were performed with the same procedures as the first period in the other experiments. Before the second period, there was given a false-instruction that a very unpleasant ES was to be given immediately after the response to some red signals. Although no ES was in fact given, this period can also be regarded as the anxiety inducing situation. The main results were as follows: (1) a prolongation of RT in the anxiety situation was demonstrated. (2) It was more pronounced in the high level of anxiety. (3) The same effect was shown also in the anxiety situation induced by the false-instruction. (4) The prolonging of RT to the reinforcement stimulus (red) can be considered as the specific anxiety effect of the conditioning. (5) The effect of the backward conditioning on RT was scarcely evident. (6) The report by Bersh and Garvin that the RT slowdown was eliminated by the use of the ready signal was not confirmed by the present experiments.
The present experiment was so designed as to test the hypothesis that the intradimensional shifts during optional shifts would be facilitated by verbal responses to the discriminanda as well as overtraining trials in initial discrimination task. A 2×2×3 factorial design was used, which incorporated two (color and form) dimensions, verbal and nonverbal responses, and 3 levels of initial overtraining. The Ss were 192 kindergarten children of the mean age of five years and three months, who were assigned to any of 12 cells equated in age and sex. They were given two-dimensional, simultaneous discrimination tasks as shown in Fig. 1. Throughout the experiment, half of the Ss was required to respond to discriminanda using verbal expression (verbalization group), while the remaining Ss were instructed to point out stimulus cards with fingers (nonverbalization group). The correct responses were informed by saying “Hit”, and the incorrect responses by “Miss”. Initial discrimination tasks were learned up to three training levels respectively: the criterion of nine correct out of 10 successive trials (no overtraining), additional 10 overtraining trials after the criterion, and additional 30 overtraining trials after the criterion. Then, the Ss were trained on an optional shift task to nine correct out of 10 successive trials. The test discrimination task consisting of 18 stimulus cards was presented in order to determine whether the optional shift had been intradimensional or extradimensional. The main findings were that (a) there was no significant difference in initial discrimination performance between verbalization and nonvervalization groups (Table 1), (b) intradimensional shift was facilitated in terms of verbal responses and overtraining, but the overtraining effect was somewhat due to the performance of verbalization group (Tables 4 and 5), (c) The order of the percentage of intradimensionally shifted Ss was relevant verbalization group, nonverbalization group, and irrelevant verbalization group, (Tables 5 and 6), and (d) faster learners of initial discrimination tasks performed intradimensional shifts more easily than slower learners. The results were discussed in relation to previous studies, mediating response theories, and dimensional preferences.
It has been shown that when a stimulus figure exists in the visual field, the figure exhibits the so-called figure-effect in the form of changes in the light threshold for a small luminous point presented at places near the figure on the same front parallel plane. The finding leads to the following possibility: An effect similar to this will also occur in the depth of the visual space. For this reason, in the present study, several experiments were performed with the following two objects in view. The first object is to examine whether such an effect can be shown to exist at all. The second is to examine what tendency the effect has, if it does exist. The experimental conditions and the method of the measurement were as follows: The stimulus objects consisted of two parallel lines which were lit to form a pair of light stimuli, with three kinds of interior interval distance with visual angles of 12′, 18′, and 48′. They were constantly presented on the front parallel plane at a distance of 592.3mm from the observer. In one experiment, only one of them was presented. When the figure was presented, a small luminous point was simultaneously placed at various positions in front and the rear of the figure as shown in Fig. 2 (A), and light threshold for the point, value t, was measured. Then the threshold for the same point was measured again at the same places when the figure was not presented, as shown in Fig. 2 (B). This threshold, called t0, is a criterion value to check the degree of the elevation of the threshold t. That is, the figure-effect was examined by the difference between t and t0, and the value of (t-t0)/t0. The results obtained in the experiments were as follows: (1) It was found that the threshold values, as a whole, were higher in the depth than in the same frontal plane to the figure. (2) The threshold value was gradually elevated as the distance between the figure and the point was increased, and at a certain place far from the figure, a marked elevation was seen. Such a tendency of the elevation was observed both in the front and the rear visual spaces of the figure. (3) Such an elevation of the threshold with the increase of the distance were more pronounced as the interval distance between the two parallel lines became shorter. In view of the above results, it is concluded that the figure-effects are present also in the depth, i.e., the third dimension, of visual space, and that they also cause the elevation of the threshold as the distance from the stimulus figure increases. Such a tendency seems to be dependent upon prevailing stimulus configuration.