Constraints on Pavlovian aversive conditioning: Implications for avoidance learning in the rat

Four experiments investigated the relative degree of control exerted by several conditional stimuli when each stimulus: (a) preceded shock, (b) followed shock, and preceded a shock-free interval, (c) was independent of shock, or (d) was novel. When the onset of an auditory stimulus had preceded shock (Experiments 1, 2A, and 3), it always exerted conditioned excitatory control. When the same stimulus preceded a shock-free interval, it never exerted conditioned inhibitory control. When the onset of light (Experiments 2B, 4), silence (2C), or darkness (1, 2D) preceded a shock-free interval, it always exerted conditioned inhibitory control. However, when one of the latter stimuli preceded shock, it failed to exert excitatory control. Several models of this phenomenon were tested. The implications of these results for the species-specific defence reaction and two-factor theory accounts of avoidance learning were outlined.     

A new apparatus for the study of avoidance conditioning in fishes

An apparatus for the study of avoidance conditioning in fishes is described. The chamber is cylindrical in shape, with shocking electrodes placed above and below the animals, and response is defined as swimming a predetermined distance in either direction along a circumferential path. This apparatus has several advantages over the conventional shuttlebox: (1)There is little constraint on the direction of swimming; (2) the magnitude of response (swimming distance) required for avoidance can easily be varied over a wide range; and (3) variation in the effectiveness of shock with the position of the animal relative to the position of the electrodes is minimized. Some sample data obtained in a free-operant experiment with goldfish are presented.     

An electrode hookup for eyelid conditioning

A method for recording eyelid CRs, using skin electrodes rather than the usual mechanical connection to the S’s eyelid, is described. Data are presented to show the comparability of the two recording systems.     

Seven-minute rhythms in stable avoidance conditioning

Investigation of the “inextinguishable” locomotor avoidance reflex revealed periodic oscillating of reaction times, motor reaction times, and latent periods, when the conditional stimuli were applied every half minute. These rhythmic properties were verified by autocorrelation. Frequency distribution of all significant wavelengths occurring in rhythmic processes of stabilized conditioning showed seven-minute wavelengths for the most part and smaller numbers of wavelengths of 5 to 9 and 14 minutes. A significant number of wavelengths of 3.5 minutes was not seen. Locomotor avoidance conditioning permitted separate automatic registration of reaction times and motor reaction times. The two values showed synchronous periodic courses in stabilized conditioning, with the motor reaction times being longer. The periodicities found, thought to be due to controlling processes in the central nervous system, can be explained in either or both of two ways: the periodic courses represent a frequency range of a structured time mechanism in the sense of a “clock inherent in an organism”; the rhythms of conditional processes represent an adaptation mechanism in organism-environment interrelations. It can be assumed that the two mechnanisms work synergistically.     

A comparison of escape and avoidance conditioning in wild and domesticated rats

In the first of two experiments, three cotton rats (Sigmodon hispidus) and three albino rats were exposed to instrumental escape, unsignaled avoidance, and signaled avoidance, in that order. All subjects learned the escape procedure quickly, with the albino rats having generally shorter latencies, higher response rates, and requiring fewer sessions to reach the criterion. When the avoidance contingency was introduced, the cotton rats continued to respond almost entirely in the presence of the shock, whereas the albino rats responded in its absence, thus displaying effective avoidance behavior. Introduction of a pre-aversive stimulus did not improve the performance of the cotton rats. In the second experiment, five cotton rats and four albino rats were exposed to a free-operant (Sidman) avoidance procedure with a shock-shock interval of 3 sec and a response-shock interval of 20 sec. The cotton rats initiated responding at lower shock intensities than the albino rats, but their asymptotic avoidance responding was far less effective.     

The role of the cs and avoidance period duration in discriminated avoidance conditioning

In an experiment investigating the effect of CS duration on discriminated bar-press conditioning, subjects were assigned to one warning period duration (1.5, 5.0 or 15.0 sec.) on the first day of avoidance training, and to one of the three durations on the second day of training. On each day avoidance behaviour was greatly influenced by the duration of the CS (warning) period, but the duration on day I had no effect on the second day's avoidance performance. Groups receiving no CS during training, although provided an avoidance contingency, showed little conditioning, but produced highly significant amounts of intertrial responding. In a second experiment, subjects receiving CS on day I were shifted to no CS on day II. Avoidance performance on day II was not significantly different from the day II performance of subjects in Experiment I having two training sessions with the CS present or absent on both days.