The next major step in the history of experimental neurosis took place in the laboratory of Martin E. P. Seligman, who, with his coworkers, discovered that dogs exposed to traumatic inescapable shock showed signs of neurotic elaboration and disintegration on cognitive, emotional, and motivational levels of organization (Seligman and Maier, 1967; Maier et al., 1969).
Subjects were small dogs of unknown origin that were divided into three groups: escape trained (ET), yoked control (YC), and control (C). The ET group was exposed to escape training involving shock applied to the foot pads of the hind feet while restrained in a Pavlovian hammock. Flat panels located on either side of the dog's head would immediately terminate shock if pressed by side-to-side movement of the dog's head. The YC group, which was simultaneously exposed to identical conditions and stimulation but was not able to escape shock by moving the panels, was exposed to 64 traumatic inescapable shocks occurring every 90 seconds (average) for a duration dependent on the speed and pattern of ET's escape responding, for a total of 226 seconds of shock overall. The ET group could terminate the traumatic shock with the appropriate press of the fitted panels. The C group received no escape training. The following day, all three groups were exposed to escape-avoidance training in the presence of a visual CS (turning off of lights in a conditioning compartment) 10 seconds continuously prior to the delivery of shock. The required avoidance response was jumping over a hurdle (adjusted to the height of the dog's shoulder) into an identical adjacent compartment. A dog that jumped as the light was turned off could avoid shock altogether. All subjects were exposed to 10 trials of escape-avoidance conditioning.
Both the ET and C groups readily learned the shuttle-box avoidance response. The YC (helpless) group, however, exhibited great difficulty in mastering the required behavior. Most of the helpless dogs failed to escape shock by jumping over the barrier when tested. Instead of making an effort to jump, they displayed intense panic reactions followed by impassivity—they simply laid down and whimpered on a wire grid of pulsating shock. As testing proceeded, they made no effort to escape at all. A striking outcome of Seligman's experiment was that inescapable shock had dramatic negative and interfering effects on postshock learning. Even when helpless dogs occasionally succeeded in jumping over the barrier, they were unable to benefit from these successes on subsequent trials. Besides failing to initiate purposeful efforts to escape shock and learning from their experience, Seligman described several other prominent characteristics associated with learned helplessness: (1) time course (after a single exposure to uncontrollable shock, most dogs recovered within 24 hours but failed to recover after repeated exposure), (2) lowered competitiveness (aggressiveness) and general vitality; (3) development of a negative cognitive set (a belief that nothing can be done, i.e., environmental events are independent of action), and (4) loss of appetite (an outcome also associated with pathological stress).
Seligman theorized that the disruption of escape-avoidance learning and collateral symptoms of helplessness were caused by the affected animal's lack of voluntary control over the traumatic event rather than the event itself. Although trauma is a necessary condition for helplessness to occur, it is causally insufficient in itself to produce the effect. Both ET and helpless dogs received identical treatment, except that the ET dogs were shocked under conditions that they could control. For learned helplessness to occur, the event must be both traumatic and outside the subject's control. Subsequent experiments with a variety of animal species have uniformly supported Seligman's conclusions. The theory has enjoyed widespread acceptance and represents a leading animal model for reactive depression and, more recently, post-traumatic stress disorder (PTSD) (Van der Kolk et al., 1985: Foa et al., 1992).
Seligman originally believed that learned helplessness was a transient effect, with recovery occurring within 24 to 48 hours. Two exceptions conflicted with this general observation: (1) dogs receiving multiple sessions of inescapable shock exhibited protracted signs of helplessness, and (2) animals raised under laboratory conditions tended not to recover from the helplessness effect. He speculated that the likely cause for this difference could be attributed to past experiences with controllable trauma. Laboratory-reared dogs were more naive (never having been exposed to es-capable traumatic events) than the dogs of "unknown" origins that he had used. The latter group had come from backgrounds that may have included exposure to escapable traumatic handling. To clarify the effects of past exposure to controllable shock, Seligman performed a series of experiments on rats and found that naive adult rats did not recover over time from the effects of inescapable shock. Another group was trained to escape shock and then exposed to inescapable shock. Previous exposure to escapable shock appears to have immunized the escape group against the effects of learned helplessness (Seligman, 1975). Helplessness studies on weanling rats exposed to inescapable shock have demonstrated persistent interference effects lasting into adulthood. Weanling rats exposed to the immunizing effect of escapable shock did not exhibit learned helplessness when exposed as adults to inescapable shock. In fact, immunized rats did slightly better on escape-avoidance tasks as adults than did nonshocked controls (Hannum et al., 1976).
Reversing the helplessness effect was possible only by physically forcing the dogs over the shuttle-box barrier. Dogs had to be physically prompted to jump over the barrier for as many as 20 to 50 trials before they began responding on their own. After directive exposure was carried out, helpless dogs began responding like normal ones (Seligman et al., 1968).
Family dogs habitually exposed to unpredictable/uncontrollable punishment are at risk of developing disturbances associated with the learned-helplessness disorder. Traumatic punitive events involving excessive startle reactions or physical pain, which are poorly coordinated with identifiable avoidance cues or response options, meet the operational criteria of inescapable trauma. The occurrence of such interaction is particularly common in cases where punishment takes place long after the event, or when it is applied out of anger. Under these conditions, the owner should be careful not to punish but to think through a plan based on sound behavioral practice to change the offending behavior.
Another key consideration is to avoid the application of traumatic or highly threatening punishment altogether. Dogs exposed to excessive punishment will never reach their full potential but rather are bound to grow gradually callous to their owner's abusive treatment, appearing not to feel punishment by their lack of responsiveness it. in fact, helpless dogs appear to develop an endorphin-mediated analgesia stimulated by uncontrollable trauma (Drugan et al., 1985). On a cognitive level, helpless dogs have simply learned to take punishment but not to benefit from it. They have fallen victim to a negative learning set that prevents them from responding appropriately under compulsion, perhaps believing that anything they might attempt to do will only fail.
More recently, Sonoda and colleagues (1991) demonstrated that interference effects paralleling those of learned helplessness can be obtained under conditions of uncontrollable appetitive training involving the non-contingent acquisition of food. The interference effects observed adversely affected cross-motivational learning involving shock-escape training. They performed a series of experiments with rats in which three groups were exposed to various conditions of control or loss of control over the acquisition of food. Initially, all of the rats were exposed to a continuous schedule of reinforcement for lever pressing. The rats were then divided into three experimental groups. one group was exposed to additional training under both an FR 5 and then, on the following day, an FR 20 schedule of reinforcement. A second group (the loss-of-controllability group) was yoked to the first group, so that these rats received food on a schedule independently of what they did or did not do with respect to lever pressing. Finally, the third group was given the same number of pellets earned by the first and second groups but en masse in their cage. The various rats were then exposed to a simple escape-training situation (shuttle box) in which they had to jump over a barrier to escape shock. interestingly, the rats exposed to the loss of controllability contingency proved unable to learn the shuttle-escape task. This result is consistent with the cognitive interference effects exhibited by dogs exposed to uncontrollable shock:
The important difference between the contingent rats and the loss-of-controllability rats is whether or not a food outcome occurred when no target response was given. Food never occurred for the response-contingent rats when no target response was given, whereas food occurred when the loss-of-controllability rats lost the contingency between a target response and a food outcome. Therefore, the loss-of-control-lability rats lost the contingency between a target response and a food outcome. Hence, the interference effects in the present two experiments suggest that the cognition of the contingency between a response and an outcome is an important factor in governing an organism's behavior. (Sonoda et al., 1991:274)
Was this article helpful?