This investigation studies the changes produced in the electroencephalogram during the course of conditioning, i.e., from the beginning of conditioning to the end of extinction. Pure tone (1, 000 cps and about 70 db) and light were used as conditioned and unconditioned stimuli respectively. The conditioned response Could be traced in EEG in almost all the subjects. The characteristics of the changes in EEG found in this experiment may be summarized as follows : (1) Under tone stimulation alpha waves do not show any tendency to increase with the progress of conditioning. (2) In the early stage of conditioning and also at the end of extinction, alpha waves gradually increase during tone stimulation, but this tendency to increase takes a different form in each case. (3) On the other hand, when the conditioning is well advanced and comes to be nearly completed, alpha waves appear very much in the begining of tone stimulation and decrease gradually with its continuance. This decrease may be considered as the manifestation of the expectant reflex and perhaps is characteristic of the conditioned response in EEG. (4) With respect to the duration of the disappearance of alpha waves, the extinction process seems to have a special pattern of its own. (5) There is a limit to the duration of the disappearance of alpha waves. even if the number of conditioning is increased ; the duration is limited by the duration of light stimulation. (6) In the first extinction trial, after the cessation of tone stimulus alpha waves increase in inverse ratio to the number of conditioning. Alpha waves disappear, however, after ten trials of conditioning. (7) The number of appearance of the conditioned response is approximately proportional to that of conditioning. From (6) and (7), it may be inferred that the greater the number of conditioning is, the greater will be the strength of conditioning. (8) The changes in EEG produced after light stimulation in conditioning bear a resemblance to those that appear after the conditioned response in extinction. In both cases, alpha waves disappear for the first one or two seconds and increase rapidiy for the next five to seven seconds, and then they decrease again. The process continues for sometime in a smaller scale.
In a situation where extinction of a conditioned (continuously reinforced) response would be indicated by the change to an alternative response for the same goal, the effect of “relative” effortfulness of the conditioned response to that of the alternative response upon extinction and spontaneous recovery was examined. Hodologically, in this situation, the occurrence of the alternative response would mean that the psychological force has changed its direction. In view of the fact that extinction is a function of the effortfulness of a response, it was expected that the greater the relative effortfulness the smaller the resistance to extinction. No significant difference, however, was found between the effects of different relative amounts of effort upon extinction, in spite of the fact that they were accompanied by statistically significant difference in the tendency to spontaneousr recovery : the greater the relative effortfulness of a response the smaller the tendency to spontaneous recovery. These results seems to indicate that spontaneous recovery is more sensitive to the relative effortfulness of a response than extinction. No significant decrease in habit strength as measured by the number of errors, latency, and time required to perform the response was found even right up to time of complete extinction. Extinction, therefore, seems to be due to the changed direction of a psychological force rather than to the decrease in habit strength.
In his previous study (1) the writer confirmed the appearance of the S-effect upon the skin, and pointed out the fact that the visual image accompanied this phenomenon in the normal seeings. The present investigation was designed to see whether of not the vision was an indispensable requisite for the occurrence of S-effect upon the skin. Three momentary pressure stimuli were presented in succession upon forearms and the subjegts were asked to compare the length of time interval between the second (R2) and the third stimuli (R3) with that between the first (R1) and the second stimuli. The following three types of persons served as the subjects : (1) the four normol seeings (adults) (n.), (2) the two partial seeings and the two total blinds from their boyhoods (high school students) (p. s. and 1, t. b.), (3) the three total blinds from their early childhood who are comparable to the congenital total blinds (high school stuqents) (e.t.b.). All subjects were required to keep their eyes closed during experiment. Measuring eight times by the method “Das Verfahren des vollen Ab- und Aufsteigs”, the equivalents were compuited. The main results were as follows : 1. Three stimuli were presented upon the right forearm as shown in Fig. 1. If the spatial distance between R2 and R3 was longer (shorter) than that between R1 and R2, then time interval between R2 and R3 was overestimated (underestimated) by n. and p.s. and l.t.b. On the other hand if both distances were equal, both time intervals appeared approximately equal for them. But such a consistent effect did not occur in e.t.b. (Exp. 1) (cf. Table 1 and Fig. 2) 2. The first and the second stimuli were presented upon the right forearm and the third stimulus was presented upon the left forearm as in Fig. 3. If the spatial distance between R2 and R3 was longer (shorter) than that between R1 and R2, then time interval between R2 and R3 was overestimated (underestimated) by n. and p.s. and l.t.b. But this S-effect did not occur in e.t.b. (Exp. 2) (cf. Table 2) 3. The first and the third stimuli were presented upon the right forearm and the second stimulus was presented upon the left forearm as in Figs. 4 and 5. Under the condition of Fig. 4 an illusion of spatial distarice appeared, that is, the distance between R2 and R3 appeared longer than that between R1 and R2 : moreover, time interval between R2 and R3 was overestimated by n. and p.s. and l.t.b. Under the condition of Fig. 5 the former distance appeared shorter than the latter ; besides, time interval between R2 and R3 was underestimated by them. On the other hand, such an illusion of distance and S-effect as mentioned above did not take place in e.t.b. (Exp. 3) (cf. Table 3) Thus it has been found that the S-effect appears in the partial seeing and the total blind from his boyhood as in the normal seeing, and that the manner of visual interference in recognizing temporal length in the former is the same as that in the latter ; besides, the S-effect does not appear in e.t.b. Taking the above-mentioned facts into consideration, it may be concluded that the vision is an indispensable requisite for the occurrence of S-effect upon the skin.
This experiment has been conducted to investigate the relationship between the cff and the experimental variables of intensity of illumination, sex, age, and individual variability. An electronic flicker unit was employed to produce and to control the flicker light, and two Kodak-Wratten filters with 0.3 and 1.0 density respectively were used to produce three levels of intensity of illumination. Forty subjects were used : ten males and ten females with the range of age from 22 to 26 years, and ten males and ten females in the 27 to 31 years age group. The actual experimentation consisted of two sessions with three minutes intermission between them. For each session six readings were taken, three in the ascending order and three in the descending order. This procedure was repeated for each of the three levels of illumination which were presented according to the Latin square design. The dominant eye of the subject. was used throughout the experiment. Summing up the findings of this study, the following tentative conclusions were reached : 1. The cff and the logarithm of intensity of illumination have a linear relationship. 2. The cff value is higher for male than. for female. 3. The older group had a lower cff valuesthan the younger group but the differencewas not significant. 4. The intervariability of the cff tends to increase with intensity of illumination while the intravariability of the cff tends to decrease with intensity of illumination. 5. In general both intervariability and intravariability tend to decrease with age. 6. In general, the cff value is constant from session to session.
In this paper the relation between character measured by Rorschach test and the degree of physical maturation was duly considered with the result that the concrete structure of character which maturation prescribes was clarified. Those who were tested were 156 boys and 166 girls ranging from 2nd grades to 8th grades. The test of character was made by the use of 10 Hans Huber plates, the physical maturation being decided with the X-ray pictures of the carpal bones. The results of character test tell that the fields upon which physical maturation has clearly influence are the number of responses, good form responses, movement responses, apperceptive type and W. answers of good form, while the fields upon which it has pretty influence are whole responses computed from plates and quality, colour responses, experience type and animal responses. The qualities of character differentiating groups of higher and lower maturation (H-group and L-group) which can be under-stood from these results will be summarized as follows : (1) The differences connected with undulations of intellectual faculties and peculiarity of intellectual activities were revealed in diverse results of test with distinct contrast. The H-group showed rich imagination, quick and ardent association activities with clear and reasonable process, high competence for summarization and systematization, sufficient abilities for abstract thinking, high abilities to seize actuality, excellence in personal responses and abundant originality. It follows, therefore, that those in H-group have generally high intellectual faculties. L-group showed lower levels in every respect compared with the features mentioned above. In addition to that, poverty of idea, stiffness in mental activity, crudeness and superficiality in the way of thinking, together with imitative tendency were comparatively conspicuous in them. (2) With respect to the features mentioned above, H-group showed, when confronted with difficulties, proper reasonable adaptability with their strong thinking power which means that they have high intellectual and social adaptability, while L-group failed in those intellectual and social adaptability, unreasonableness or inability of adaptation being their general tendency. (3) H-group showed introversive tendency, while L-group showed comparatively high extrotensive tendency. (4) H-group tended to fall into depressive mood, but to have stabilized emotion, whereas L-group which had the rather tendency of self-centricity tended to run into elated mood, low emotional adaptability and emotional immaturity with little stabilization in emotion, frequently showing impulsive outbursts etc. (5) In those respects mentioned above, some sex differences were observed, but, in general, common tendencies or features were more conspicuously observed.
Generally, resistance to extinction is greater after partial reinforcement procedure than after continuous reinforcement. This phenomenon may be explicable mainly by F. D. Sheffield & Temmer's generalization decrement theory (2) and V. F. Sheffield's after-effect theory (3). The former emphasizes the changes of cue pattern from training period to extinction period. due to the elimination of reinforcement. Since these cue changes are smaller in the groups of less reinforcement rate, more persistent resistance is expected in these groups. The latter pays attention to the facts that in partial reinforcement procedures response is also conditioned to the cues involving after-effects of non-reinforcement. Therefore, so far as moderate number of non-reinforcement trials are concerned, the conditioning to such non-reinforcement cues are stronger in the groups of greater reinforcement rate, and greater resistance is expected in these groups. Practically, however, the most favorable rate will be expected in the middle of these two extreme rates. Our experiment was performed to test the above considerations. Rats in Group-1/1 were placed on electrified grill of grey compartment in every trial and allowed to escape from shock by running to the adjacent non-electrified white compartment (escape procedure). Group-1/2 received escape procedure as Group-1/1 in half of the total 40 trials, but the electric shock was eliminated from both compartments in the remaining half trials which were interspersed semi-randomly with shock trials. Group-1/4 received escape procedure in a quarter of trials and was given no shock ni the remaining trials. The results obtained were as follows : 1) In the non-shock trials in training process, more speedy runnings were performed by Group-1/2 than by the Group-1/4. However, the difference did not reach a statistically significant level (Table 1). 2) The order of running speed in the early extinction stages was : Group-1/2, Group-1/4, Group-1/1. (Fig. 2-c). The differences between Group-1/1 and Partial Groups were significant, and difference between Group-1/2, and Group-1/4 was nearly significant (Table 2). With respect to the number of trials to reach the extinction criterion, the order of three groups was the same as in the case of running speed. The differences between Group-1/1 and Partial Groups were significant or nearly significant, but there was little difference between Group-1/2, and Group-1/4, (Table 2). The following conclusions were deduced from these results : 1) The faster extinction in Group-1/1 is attributed mainly to the greater cue change due to elimination of electric shock as a cue. However, one should understand that this cue change is different from change of cue pattern due to elimination of rewards as in reward training situation. 2) Though Group-1/2, performed more speedy running than Group-1/4, in early extinction stage, number of trials to reach extinction criterion was almost equal in both groups. This seems to indicate that the less reinforcement rate in Group-1/4, was favorable to persistence of response.