Compared to lever-press avoidance learning, easy and quick learning is expected when a running response which is a part of the “Species-specific defense reactions (SSDRs)” (1) i.e., fleeing, is used as an avoidance response (4, 5). However, even in the running-wheel situation, some rats were discarded due to failure to respond (2). Rats might avoid a shock by crouching to the same pole of the grid, because a nonscrambled shock generated by a slipring commutator was used. This suggested the possibility of promoting avoidance learning if a scrambled shock is introduced. However, such studies using a running response in a rotating cage have been little attempted, because of the complexity of the mechanism to transmit a scrambled shock to the rotating grid.
Recently, ISO (3) attempted to improve the rotating cage, and reported effective Sidman-avoidance learnings in rats. A newly designed shock-transmitter mechanism was attached to the cage, and the scrambled shock was delivered on the grid. In the three experiments, response-shock (R-S) interval was 20 sec, shock-shock (S-S) interval was 5 sec, a shock duration was 0.5 sec and the shock intensity was 200 V AC. Then, in experiment 1, the magnitude of rotation to be counted as a response was tested in three times of 1 hr sessions. Rats learned more quickly in a quarter turn condition than a half condition. In experiment 2, the maintenance of learning over sessions was tested. Rats showed some superior maintenance of avoidance response without over-night decrement for consecutive 8 days. In experiment 3, the continuation of avoidance responses through a long session was examined. Avoidance responses were continued until the end of the session over 6 hours, and rats ran a mean of 1544 m with a mean of 42 shocks during 6 hours. Such results indicated the utility of the rotating cage for the avoidance learning in rats.
Then, in the present experiment, the effects of shock intensity and R-S interval which are the basic variables to affect the avoidance learning, on Sidman-avoidance learning in rats are tested using the same apparatus as ISO (3).
Sidman-avoidance learning of Wistar rats as a function of shock intensities (100, 200 and 300 V AC) and response-shock (R-S) intervals (10 and 20 sec) was examined in a newly designed rotating cage [ISO (3)]. The shock-shock (S-S) interval was fixed at 5 sec with a shock duration of 0.5 sec. One habituation, two avoidance trainings and one extinction were continued for one hour each. Such sessions were split weekly. All the rats learned avoidance responses easily, especially well in the condition of 20 sec R-S interval with a shock intensity of 200 V.

View full abstract

At 20 days of age, thirty-six ICR/JCL male mice were randomly assigned to one of the following housing conditions ; in isolation (N = 18) and in groups of three (N =18). Following 45± 5 days of confinement in each housing condition, animals were placed singly and left for an hour in one of two compartments of an observation cage divided by a partition. On the next day subjects were again placed in the same way as the previous day. Ten minutes later, the partition was removed and the animal was permitted access to the entire cage. The behavior of each animal was recorded for twenty minutes with the aid of VTR system. The recorded behavior was described by the sequential sampling method on the basis of 15 items of behavior (Table 1). In order to examine possible “dyadic” structures, the first-order transition analysis and the single-link cluster analysis were applied to the first 10-min records of behavior.
Two clusters were found in the behavior (Fig. 1); one including the behavioral items accompanied with “stretching” (cluster I) and the other not (cluster II). The internal structures of cluster I were identical between the two housing conditions, but the structure of cluster II in isolated animals differed from that in grouped ones, especially with regard to the partial structure around “rearing”, “pausing”, and “face-washing” (Fig. 2 and 3). An analysis of transition probabilities in and between the two clusters suggested that cluster I was the initial and transient component of the sequence of behavior (Fig. 2 and 3). This was confirmed by the fact that the number of individuals showing the items with “stretching” markedly decreased approximately within a minute, particularly in grouped animals (Fig. 4). However, in isolated animals a small amount of “stretching” sporadically occurred during the entire observation period.

View full abstract

[in Japanese]

View full abstract