It is known that the Pavlovian internal inhibition of a conditioned response possesses and inhibitory potential on other S-R connections. However, whether the inhibition produced by experimental extinction has the same inhibitory power as that of internal inhibition is open to questions, because the inhibition by extiction differs from the inhibition of other types in several respects, for example, only the former shows spontaneous recovery. The present experiment was undertaken to answer this question and find further partcularities of the inhibitory process by extinction if any. Twelve rats were trained in bar-pressing upon the presentation of either light or sound in the Skinner box. After attainment of criterion, the response connected to the sound stimulus was subjected to exinction. Reaching the extinction criterion, i. e., 5 successive non-responses, the rats were divided into two groups, experimental and control. Immediately after grouping, test-extinction for the light stimulus was run, in which the light stimulus was presented alone for the control group, whereas the light was accompanied by the sound for the experimental group. Next day, the same procedure was repeated for both groups, i. e., the 2nd extinction and the 2nd test-extinctions. Result : The mean number of responses and the mean latencies per trial were shown in Fig. 1 and 2, and Table 1 gives the result of t-test based on those values. An insignificant difference between two groups during acquisition and the first extinction indicated that the two groups were homogeneous. Moreover, two groups were not significantly different from each other in test-extiction. This seemed to mean that the stimulus to which extinction was performed did not accelerate other extinction. But the test-extinction was faster than the preceding extinction in both groups. It was concluded that the stimulus involved in extinction would not have a power to inhibit other S-R connections, and that the tendency of non-responding built in extinction simply transferred to other conditioned responses.
The “spread” hypothesis proposed by Thorndike to explain the “spread of effect”, states that the influence of a rewarding state of affairs acts not only on the connection to which it belongs but on adjacent connections around the rewarded connection as a function of distance from reward. In an orthodox Thorndikian design of spread of effect, the effect is found to act to strengthen punished connections or more precisely to repeat error responses in the neighborhood of the rewarded one. However, just the opposite result would be predicted strictly following the Thorndike hypothesis, in that if his hypothesis is true, S-wrong R connections around reward will be strengthened and thus the “wrong” R's will not be repeated on the following trial in comparison with those away from reward. A clear cut test of the “spread” hypothesis will be made if S-R connections near reward are not punished and number of repetitions of correct responses rather than errors is calculated. Several such studies were attempted with rats (5, 11, 12, 13, 15) and one with human (6) although their own objectives were different from ours. However, these works have been criticized because of a number of uncontrolled and/ or confounding factors. In the present experiment, 12 pairs (S-R's) of two syllables Japanese nonsense letters were presented individually to S every three seconds. Only one of the 12 such pairs which is called the key item, was written in red. The experimental Ss were required to the key item with special attention among the rest items. A recall test was given immediately after presentation of the 12 paired-associates but the ordering of the stimulus letters in recall test was randomly changed from trial to trial in order to rule out response-response learning. Five recall tests were administered per S. The control Ss with 12 homogenious paired-associates, i. e., without the red item, were used to control serial position effects in the experimental results. As the key item was expected to recall better and thus to be “rewarded” more often that the rest, a Thorndikian gradient in percent recalls should appear if the “spread” hypothesis is correct. The results shown in terms of the group difference (Table 1, Fig. 1 in Text) failed to prove the prediction but the correct recall score appeared to be greater for the response immediately preceding the key item than immediately following it. The trend was enhanced not only with trials (Table 3, Fig. 2) but also when the key item was correctly recalled (Table 4, Fig. 3). The results were explained in terms of increased motivation resulting from the anticipation of the key item with respect to the better recall of the first-before item and the decreased motivation as well as the rehearsal-distraction effect suggested by Jenkins & Sheffield, with respect to the first-after response. Similar results (Table 5) were obtained by the group experiments (Groups AE & AC) where the almost identical procedures as in the previous individual test were used but no observable effects were obtained by Groups BE & BC in which the ordering of the stimulus items in recall test was not varied.
In the previous paper (2) the author investigated the relation between the tendency to reproduce an adjacent digit when the stimulus digit was a single unit and when the digits were composed of compounds as in the multiplication table. Experimental results showed that the majority of errors in multiplication was caused by the replacement of one of the component numbers by an adjacent one, viz., the replacement resulted from a reproductive tendency of adjacent digits to arise in response to the stimulus digit. The purpose of the present investigation was to confirm whether a similar result would be obtained in the adding operation of simple numbers or not. Various combinations of simple numbers were used as test-stimuli, for example, 2+5, 6+3, etc. Ss were 279 primary school pupils. In the experiment the stimulus pair-digits were presented individually to the S and he was asked to respond with the right answer as quickly as he could. The numerical intervals between the right numbers and the response ones were computed with special attention to the frequency of occurence of each interval. The results were as follows : Figure 2 in the Japanese Text shows that in the adding operation the strength of remote association as indicated by the frequency of errors declines with increasing degree of remoteness of the two digits-the right and the erroneous responses. The same characteristic was also seen in the free reproduction when single digits were used as stimuli (Fig. 1). Therefore, it was concluded that the cause of errors in adding resulted from the tendency of remote reproduction associated with the use of single digits as stimuli. The results shown in shown in terms of the frequency score difference between the remote reproduction for single and that for the adding operation reveals that the spread of reproduction or facilitation occurs when the interval between the right number and the erroneous one is small. The facilitative effect decreases as the distances between the two terms becomes larger, until finally the inhibitive effect appears. This finding is shown graphically in Fig. 5. (Failure of facilitative effect in the plus side of abscissa in Fig. 5 was supposed to be due to experimental errors.) Now, it must be noted that in the adding operation one of the two combined digits was often omitted, for example, 3+4=4. Concerning the omission it was found that the smaller digits were more apt to be dropped than the larger ones. This was interpreted to be caused by the predominant reproductive tendency of the larger digits as compared with the smaller ones, resulting in repression of the reproduction of the latter by the former. Finally, it must be noted that the preceding terms (digits to be added) of the combined digits were more liable to omission than the succeeding ones (digits to add). This is probably, due to the retroactive inhibitory effect of the latter upon the former. In concluding the author would emphasize that the practice of addition in arithmetic consists in the formation of a strong associative connection between the stimulus digits and the response digits, for example, giving the quick response 8, when S is presented with the stimulus “3+5”. An analysis of errors in the addition, therefore, in not only valuable to teaching practice but also it becomes the key to the solution of difficult problems in the psychology of learning.
This paper reports a study on the factor of need as affecting memory trace under the condition of non-intentional remembering, with a view to approaching the clarification of the generalities of memory process in the total situation. The results were as follows: (1) It was shown that the need systems of an individual gave rise to tension systems corresponding to the need systems, and the tension systems supported the memory trace and affected the tendency to recall or recognition. Namely, it was proved that the more intense the need, it gives rise to the more intense tension system, which supports the memory trace, and a stronger tendency for recall or recognition i. e., [t (SG) =F (nG) ], [|fP. G. |= F |t (SG) |]. (2) These results were equally recognized under the condition of the repeated recall and the condition of the intentional memorizing. (3) Considering the results from the aspects of development and time passage, we could not draw a decisive conclusion as to whether there are any characteristics although some characteristics were suspected. (4) Thus, it was experimentally shown that we could not neglect the inner factors of an individual on the memory trace, and the generalities of memory process are clarified only by the total situation which has the characteristics of an organizational interdependence of the structural characteristics in the stimulus for memory and the conditions of an individual who tries to memorize the stimulus.
According to the experiments by Sonohara (1) and me (2) (3), the characteristics of the recognition of figures by kindergarten children were found as follows: 1. They chose dominantly figures of the same form rather than those of the same color. 2. The figures which are the same in the main direction were chosen dominantly. 3. Color was an important agent in the recognition so far as form is the same. 4. The reversal figures were chosen dominantly. 5. The semi-reversal figures and elementary figures which consist of the same elements but different combinations were ignored. In adults figures chosen as similar were determined by the similar combinations of three factors-color, direction and form. The developmental stages from kindergarten children's characteristics to the adluts' were found about in the third and sixth grades of primary school and about in the first grade of lower secondary school. The purposes of the present investigation through the experiments I and II are more advanced analyses of the above-mentioned results. Procedure and Subjects Four sets (two sets for each experiment) of figures used in the present experiment were reconstructed from the ones used in the previous experiments. Each set consists of the standard figure and elective figures. The subject was asked to select from the elective figures the one which has the greatest similarity to the standard. In the experiment I the number of subjects from kindergarten children to adults is 656 and in the experiment II, 624. 1. Results with children : a. The choice of the figures according to form was obvious. b. The tendency to pay attention chiefly to the upper part of the figure was found in some degree. c. The importance of the reversal figures was re-affirmed. d. The ignoring of the semi-reversal figures and the elementary ones was reaffirmed. 2. Results with adults : The choice of the figures according to the three factors-colors, direction and form-was re-affirmed. 3. The developmental stages were as follows : As to the third and the sixth grades in primary school and the first grade in lower secondary school, the results of the previous experiments were re-affirmed.
The present experiments were conducted to find out how the change in the apparent length of and aperture caused by attaching a pair of arrowheads on both ends of it, affects the phenomenal velocity of moving objects behind the aperture. Two apparatus (one for the standard and the other for the comparison) are placed in front of the subject on a table in a dark room, and illuminated by a reflector lamp. An aperture (3mm × 15cm) is horizontally opened in the middle of a white paper (30cm × 63cm) which is placed in front of each of the apparatus. Behind the aperture is a moving belt on which black rectangles (3mm × 5mm) are distributed at an interval of 5mm. Arrowheads with different angles and with 6cm sides are attached on both ends of the aperture. The belt of the standard apparatus moves with the speed of 5cm/sec, and the subject is asked to adjust the visual speed of the moving objects in the comparison-apparatus to that of the standard. Results : Exp. 1. Without attached arrowheads the decrease of the aperture length resulted in the increase of the phenomenal velocity of moving objects. Wxp. 2. With arrowheads pointing towards the center of the aperture the object was seen faster than with arrowheads pointing away from the center of the aperture. Exp. 3. The presence of the arrowheads always increased the phenomenal velocity. Exp. 4. The effect of putting two lines perpendicular to the edge of the aperture on visual velocity was smaller than those of any other cases. Exp. 5, 6. From the results based on data from the measurement of illusion caused by a pair of arrowheads, we could not recognize that the change of the aperture's apparent length caused by illusion had any effect on the phenomenal velocity. The difference in the distance from the ends of the arrowheads to the center of the aperture had no effect on the phenomenal velocity. The difference in the area of the total extensions sustained by two arrowheads had a considerable effect on the phenomenal velocity. Exp. 7. The apparent length of the aperture was shortened by the motion of the moving objects. Exp. 8, 9. The intenser the figures showed the character as a reference-point (‘an anchorage’) to the moving objects, the more the phenomenal velocity increased. Conclusion : The effect of arrowheads attached to both ends of the aperture on the phenomenal velocity, would be generally caused by the following two factors : (1) the difference in the total extension sustained by two arrowheads, and (2) the strength acting as a reference-point to the moving objects. so far as our experimental condition were concerned, the change of the aperture's apparent length by illusion did not act as a factor in the change of the phenomenal velocity.
Previous animal studies indicate that spontaneous alternation decreases with the increase of inter-response intervals but Solomon (14) failed to find the same time function in human adults. He interpreted the results in terms of the subject's conception of chance rather than basic response mechanisms. On the other hand, Wingfield (18) obtained a fairly high degree of alternation (i. e. 80%) in human adults when two heterogeneous (i. e. red and blue) stimulus objects were used. From these two studies, it is expected that (1) if lack of the time function in spontaneous alternation is due to the subject's concept of chance in adults, the same function would appear in young children who would not fully establish the chance conception and (2) if this be true, although it may be slight, the function would be enhanced with two heterogeneous objects because it increases spontaneous alternation and lastly, (3) it was designed to determine whether color or position cues are more dominant for the alternation phenomenon. In Experiment I-A a very slight but statistically signifficant time function in alternation was observed in junior grade-school children who were required individually to guess 21 times in successsion in which side (right or left) correct answer were written but with no imformation given about their guesses (Table 1, Fig. 1). A group test was made in Experiment I-B where five blocks (separated by 50 min.) of 11 guesses with an inter-response interval of five sec. were required of the subject following the procedure in Experiment I-A but this time, a circle on the left side and a cross on the left were used an stimuli. Percentage of alternation decreased significantly from five sec. to 50 min., again confirming our expectation (Table 2). This time function was higher and more distinct as predicted when two stimuli were heterogeneous, i. e., red and blue in Experiment I-C (Table 3, Fig. 1). In addition alternation was found to be negatively correlated with position preferences (Table 3 & 4). Experiment 2-A showed that the percentage of alternation was greater with red and blue objects than with two red objects (Table 5). The percentage for the heterogeneous group was 90% for the initial two responses but in decreased with responses. Again position habits were inversely related to alternation. When red and blue stimuli were presented with shifting positions alternation was reduced. In addition, positional alternation was slightly dominant over color alternation but the reverse and insignificant relationship was found between the inital two responses (Table 6). The former results are contradictory to Glanzer's theory (4) but agreed with Rothkopf and Zeaman (12).
In Bruner's Information Hypothesis theory, Adaptation Level (A. L.) is used to acount for phenomena of assimilation. The present experiments were designed to test whether these phenomena were found under experimental conditions of lifted weight. In Exp. 1, a lighter (about 200g) A. L. for white weights and a heavier (about 500g) A. L. for black weight were formed in a training series. Using Helson's comparative rating scales (Amer. J. Psychol. 1954, 321-326), subjects were required to make comparative judgments on five white weights (100, 150, 200, 250, 300g) with a white standard weight (200g), and then on five black weights (400, 450, 500, 550, 600g) with a black standard weight, five times for each weight. Then in a critical series, subjects were asked to make comparative judgments on a black standard weight (350g) with its white, subjectvely equal, weight which had been measured for each subject in a measuring series without visual cues. Likewise, a black, subjectively equal, weight was used with a white standard weight (350g). At the beginning of the critical series, an effect of contrast was found ; i. e., the black weight was judged lighter than the white one, and vice versa. But with an increas-ing number of judgments, the black weight tended to be judged heavier than the white one. (Fig 2) In Exp. II, colors of the weights were changed, a heavier (about 500g) A. L. for white weights and a lighter (about 200g) A. L. for black weights were formed in a training series with the same provedure (except changed colors) as that of Exp. 1. In the critical series the black weight was generally judged heavier than the white. (Fig. 3) In Exp. III, the natural tendency of lifted weight for different colors was ex-amined. Naive subjects with no experience of training series judged in critical series. It was found that the black weight was judged heavier than the white weight. (Fig. 4) In Exp. IV, we examined whether subjectively equal weights would show any change under training series or with the lapse of time. After a training series subjects judged in a measuring series again. No differences were found between the results on the measuring series and those of the previous one except for one subject. Assuming that we have the heavier A. L. for the black body than for the white one, the results in Exp. I could be explained in terms of contrast. We must be careful in the application of the Adaptation-Level theory becouse there are phenomena of both constrast and assimilation which are phenomenally contradictory. But it is unknown what factors would determine the turning point from contrast to assimilation.
In a previous report, the perceptual similarity or difference in 9 colors which vary in value and chroma, but constant in hue (5R), was analyzes by the method of multi-dimensional scaling. The configurations of the colors obtained from the experiment, if mapped in two-dimensional Euclidean space, were in good agreement with those defined in Munsell system. These results were also reported by Torgerson who developed this scaling method and applied it to the same problem. In this experiment, the variation in hue was introduced in addition to that employed in the previous experiment and 13 colored papers were used as stimuli Table 1). The degree of perceptual difference in these colors were converted into the spatial distances by the ratio method as follows : As the standard, colors k and i were presented in a certain distance and the difference perceived between these colors were defined to be represented by this distance. The subject was asked to express the perceived difference between colors k and j (j ≠ i) by displacing color j in such a way that the ratio of the distance k and j to that of k and i corresponds to the ratio of the perceptual difference k and j to that of k and i (Fig. 2). As the stimulus i was varied on four ways, 4 different standards k and i were employed with respect to a certain pair k and j, and, as a matter of course, the size of the unit for the differences between k and j and that of the unit for those between l and j was in general not the same. Hence, a kind of transformation [Equations (2)-(10)] was carried out in order to convert all distances into the distances based upon a single common unit and the distance matrix D was obtained. The laborious computational process of estimating the additonal constant could be got rid of this way. The data for two subjects were separately analyzed by the multi-dimensional analysis. The following results were obtained for each observer. The configurations of colors were found to lie in the three-dimensional Euclidean space in which the distance between any two colors represented the perceptual difference between them. In this psychological color space, the three attributes of color were identified as the cylindrical co-ordinate axes, i. e., the vertical axis represented the value (Fig. 4), and the radius vector orthogonal to the value axis represented the chroma (Fig. 3) and the argument represented the hue (Fig. 5). It should be noted that this co-oedinate system was empirically derived from the data themselves, not enforced from the definition of the Munsell system. The configurations of the colors agreed fairly well with the positions of these in the Munsell system as shown in Figs. 6 and 7, if the latter was superimposed upon the psychological color space in an appropriate way. The value was discovered to cotribute more to the impression of difference between colors than the chroma. These findings which were confirmed in both of the two observers of this experiment are in harmony with the results of the previous investigation which dealt with the two-dimensional color space.