The purpose of this report was to clarify the concepts of “generalization” and “differentiation” in verbal learning from the view of mediate process. The experimental design used was as follows: Each of the four conditions, ht, lt, hs and ls was learned to a criterion of 5 errorless trials [ht: each of two stimulus items (adjectives) which were high similar each other was paired with a common response item (nonsense syllable), hs: each of high-similar stimulus items was paired with each different response item, lt: each of two stimulus items which were dissimilar each other was paired with a common response item, ls: each of dissimilar stimulus items was paired with each different response item.] After the original learning (OL), each condition was immediately learned under 2 types (t′ and s′) of transfer learning (TL). (e.g., htt′ and hts′ was learned for ht in OL, in t′, two stimulus items were paired with a common response item (a letter of alphabet) and in s′, each stimulus item was paired with a different response item in paired associate learning. Experimental hypotheses: A) According to traditional uiew of “generalization” and “differentiation”, in OL, learning is easier in ht than in lt and also in ls than in hs. In TL, in the comparison of the number of correct response on the first trial, the difference between two conditions htt′ and ltt′, hst′ and lst′, hts′ and lts′, and hss′ and lss′ should not be expected. Because it may be generally considered that the performance by the criterion of 5 errorless trials brings out perfect learning. B) According to the view of mediate process, in OL, the same expectations as hypothesis A are made. But hypothesis B should be expected on the basis of mediate process on each stimulus item, that is, learning is more difficult in lt than in ht, because in lt the differential response (dr) on each stimulus item disturbs the establishment of new experimental response integration and also more difficult in hs than in ls, because the common dr on stimulus item disturbs the establishment of new differential response integration. In TL, according to hypothesis B, the transfer effects should be expected on the basis of the degree of response integration established in OL, therefore, transfer learning is easier in condition htt′ than in ltt′ and also easier in lss′ than in hss′. The main results were as follows: In OL, the results supported the expectations, and moreover a new remarkable finding was obtained, that is, learning by lt was not easier than ls, though ls had two times responses to be learned. It seems to me that traditional view of “generalization” and “differentiation” could not give the proper explanation about this finding. The most suitable explanation of the probable ones may agree with the view of mediate process. In TL, 1) In comparison of the numbers of correct responses on the first trial the results were not consistent with hypothesis A but were clearly consistent with hypothesis B. 2) In comparison of the saving score by [(OL-TL)/(OL+TL)]×100 we could obtain the remarkable finding that we did not find from the number of correct response on the first trial in TL, that is, the effect of OL in lt interfered transfer learning. This finding could not be explained from view of hypothesis A, even if we expand the concept to “semantic generalization”. The view of hypothesis B explained this finding in terms of the interference resulting from the weak response integration in OL, for in OL,
Two experiments were designed in order to determine the effect of the amount of prior learning on associative transfer and to confirm the two-factor theory of learning. In experiment I, 12Ss in each of four groups learned a list of six pairs of two-syllabled nouns (A-B), which was followed by the second re-paired A-B list (A-Br). The amount of the prior learning was varied amoung the groups by the number of correct responses; one to three correct responses (Group 1), four or five correct responses on a single trial (Group 2), one perfect trial including six correct responses (Group 3), and one perfect plus four perfect trials (Group 4). The second list was learned to the criterion of one perfect trial for all Ss. It was found that associative interference first increased and then decreased as the degree of the prior learning increased, but associative facilitation was not observed even in the overtrained group (Table 2). Maximum interference occurred when the first list had been learned to one perfect trial. However, the results seemed to be different between the early and later stages of the second learning. Two list of seven pairs of nouns were used in Experiment II. Six groups, 10Ss each, were differentiated in terms of the amount of the prior A-B learning, namely, the number of anticipation trials: one to eight trials respectively. All Ss learned the second A-Br list to the criterion of five anticipation trials. The results (Table 4) showed that associative interference increased up to the point where the first list had been learned to three anticipation trials, and then decreased as the degree of the prior learning increased. Now, a signifi cant associative facilitation was found for the highest overtrained group. The overt intrusions failed to show a consistently increasing trend. The findings were interpreted in accordance with our two-factor theory of learning; that is, specific S-R connections in the first list have negative effects on the learning of the second list, while general learning sets facilitate it. They were discussed in connection with our previous study and the relationship between the data obtained under the performance criterion (Exp. I) and those under the trial criterion (Exp. II) were considered. Other theories of learning or transfer of training were stated in relation to our theory. It was suggested that both extinction of A-B responses and acgisition of A-Br responses might be included in the process of A-Br learning.
In this paper, the effect of an experimental extinction on another response process is reported. The problem is whether the experimental extinction of a conditioned response (lever pressing in a Skinner box) improves or inhibits the learning of another response (running response in a runway). Subject: Thirty-five female and twenty-six male white rats, six to seven months of age and 192.5g of mean weight, were used. Apparatus: A corridor-type runway connected with the Yagi-type Skinner box, as shown in Fig. 1, was adopted, Dotted lines doors, the door of the S box being a guillotine door and that of the G box, a hinged door. L denotes a lever and circles, food dishes. Procedure: All rats were first trained on bar pressing (a pellet, 50mg, for each pressing) in the S box and then the doors of the S and G boxes were so opened as to make the rats adapt themselves to the whole apparatus (S, R, G, in Fig. 1). In the test the subjects were divided into four groups. As soon as the lever pressing in the S box was extinguished under four conditions mentioned in Table 1, the door of the S box was opened and the response time (latency plus running time) to the G box was daily measured for one trial for ten days. They received three pellets (150mg) in the G box. Each rat in four correspondent control groups was first detained in the S box without a lever for each assigned period mentioned in Table 2, the detained period of each group corresponding to the time required by each experimental group to reach its own criterion. The door of the S box was then opened and the response time to the G box was daily measured for one trial for ten days. Results: As shown in Fig. 2, the groups of longer extinction (G3 and G4) were slower in running response to the G box than the remaining groups. It might be due to the generalization of extinction, and the results of these groups corresponded to those of C3 and C4 which were lower in running response to the G box than the residual control groups. On the other hand, G2 in which the bar pressing was extinguished for a relatively short time tended to be even faster in running response than G1 of still shorter extinction, although the statistical difference was not significant. The running time of G2 did not correspond to that of C2 which was slower than C1. It is herein implied that there may be some mechanism in a short time extinction to facilitate the running response. As to the control groups, those which were longer detained in the S box were also slower in running response to the G box, and this might be due either to the difficulty to know when the door of the S box would be opened, or to the adaptation to the S box. As shown in Table 3, a statistical analysis of the time of running from the S to the G box indicated that there was. a significant difference among the subgroups, G1 to G4 and C1 to C4, and also between the experimental and control groups, although no significant difference was found between G1 and G2. Conclusions: A relatively short time extinction in one response neither reduces the response tendency of another response nor inhibits its learning, but rather facilitates the learning of another response. It may therefore be suggested that an increment (ΔD) in the original drive is produced by a short time extinction and ΔD×H, the effect of this increment on E (reaction potential) is larger than the inhibition made by the short time extinction. Though ΔD may also be produced by a long time extinction, ΔD×H or the effect of ΔD on E is not larger than the inhibition made by the long time extinction, as shown in Fig. 3. The inhibition mentioned above implies generalized from the bar pressing to the runway response. A strong tendency toward a running response in G2 even in the first trial when the presence of food in the
In a previous paper (Fujita, 1955), the author found that the percentage of spontaneous alternation in a T-maze with the reward at both ends decreased as the period of food deprivation increased. From this result and other related facts, the author concluded that the reinforcement seemed to be the cause of this repetition tendency. However, there is another possibility that food deprivation itself might have decreased the percent alternation. Hence the same experiment under the non-rewared condition remained to be performed to check the effect of reinforcement on alternation. Moreover, the author's paper was followed by several experiments and theories on the relationship between alternation and reinforcement, among which two (Fowler, et al, 1959; Thompson, 1960) agreed with, and two (Iwahara, 1956; Walker, 1956) opposed to the author's result. From above reasons, the main porpose of the present experiment was to find the effect of food deprivation and reward on alternation behavior to add another datum on this problem. The effect of the intertrial interval was tested too. Seventy-two rats were subjected to the following feeding and experimental schedule of five weeks. The first week was the five days of the continuous satiated period and two days of the continuous deprivated period. The same feeding schedule was repeated in the second week, and then the rat were randomly divided into three equal groups (Gr. I, II and III). On the third week, while the feeding schedule was the same as before, the test of alternation began. The Gr. I was tested on the last day of the satiation period, i.e., 0 hour of food deprivation. For the Gr. II and Gr. III, the test were after the deprivation of 24hr and 48hr, respectively. Half the Ss of each of these groups were given the food reward in both goalboxes, and half were not given. Each S runs 11 successive free choice trials per test. This schedule continued for the 4th and 5th week. Three kinds of inter-trial intervals, i.e., 0, 60 and 300sec, were used for all rats in different orders on three weekly tests. It was found that a) under non-rewarded condition, there was a tendency to increase the percent alternation with an increment of the hours of food deprivation. This tendency, however, was not significant. b) Under rewarded condition, an increment of hours of deprivation significantly decreased the frequency of the spontaneous alternation as found in the previous experiment. c) Compared with the non-rewarded condition, reward significantly decreased the percent alternation in the group of 48hr food deprivation, d) The longer the inter-trial intervals, the lower was the percent alternation in all conditions. e) The running time was significantly decreased by the increment of the hours of food deprivation in both the rewarded and non-rewarded conditions. From above results, an increment of food deprivation itself did not seemed to be the cause of the decrement of the alternation, but the interaction with the presence of reward seemed to be more determinative factor on this phenomena.