Liddell The Cornell Experiments

Howard S. Liddell (1954, 1956) studied experimental neurosis in farm animals, especially sheep and goats. Liddell's experiments involved exposing these animals to repeated simple and difficult discrimination tasks involving various stimuli and mild shocks while they were restrained in a Pavlovian frame and harness. The level of shock used by Liddell was very weak (barely perceptible to a finger moistened with salt water) but sufficient to elicit a vigorous unconditioned withdrawal response in the test animals. He utilized various conditioned stimuli ranging from somatic (a rhythmic pressure remotely applied to the skin) to auditory (metronome, bell, buzzer) and visual (dim and bright light) signals.

The typical animal was trained to perform various discrimination tasks involving conditioned stimuli predicting the presence (CS+) or absence (CS—) of shock. The behavior of one of these animals, a sheep named Robert, was described in detail by Liddell in his book Emotional Hazards in Animals and Man (1956). Robert had undergone extensive training over 3 years involving simple and difficult discrimination tasks. For example, he had been trained to respond positively to the sound of a buzzer and also to a metronome set to click once per second. The metronome or buzzer was presented 10 seconds before shock was delivered through electrodes attached to the animal's right front leg. Negative conditioned stimuli (those associated with the absence of shock) were also conditioned. Whenever the sound of a bell was presented, for example, it was never followed by shock. This sort of discrimination was easy for the sheep to master. Other negative (inhibitory) conditioned stimuli were also employed that were more difficult for Robert to differentiate from the positive or excitatory conditioned stimuli. This was especially the case with discriminations involving various metronome beats—some rates being associated with shock while others were pre dictive of its absence. These more difficult discriminations progressively resulted in the elicitation of greater distress and behavioral disturbance. Negative responses were conditioned to metronome rates of 120, 100, 92, 84, 78, and 72 clicks—all of these rates of clicking predicted the absence of impending shock. Robert easily learned to discriminate the metronome set at 60 clicks per minute (positive CS predicting shock) from the metronome set at 1 20 clicks (predicting the absence of shock). As the rate of clicking neared the positive stimulus, however, the sheep became progressively reactive and disturbed. Exposure to the 72-click rate followed a minute later by the presentation of the positive CS (metronome set at the 60-click rate) resulted in Robert failing to respond appropriately as he had done many hundreds of times in the past to the positive CS. His failure to predict the impending shock resulted in an exaggerated and inappropriate response when shock was finally delivered. Liddell describes this demonstration and the result:

Exactly one minute after metronome 72 has ceased we sound the metronome at 60 beats per minute and for the first time during the hour Robert fails dramatically in interpreting the signal most familiar to him—the signal, which, since the beginning of his training three years ago, has always meant shock. As the clicking begins at once a second he freezes with forelegs stiffly extended and with signs of respiratory distress. In fact, he duplicates his reaction to the just preceding difficult negative signal, metronome 72. At the end of 10 seconds the sound of metronome 60 is terminated by the usual shock to the right foreleg. However, Robert's reaction to this mild unconditioned stimulus is quite unusual. He leaps violently upward with both forelegs in the air but then immediately resumes a tense pose. (Liddell, 1956:10)

Liddell and his associates studied many different procedures for elaborating neurotic behavior. However, in general, the conditioning procedures he used were for the most part limited to monotonous and repetitious discrimination tasks, typically involving 20 trials a day, 5 days a week, over the course of several months. Their theoretical account of experimental neurosis emphasized the importance of monotonous repetition and restraint (i.e., loss of control) as the crucial factors involved in the production of neurotic disturbances. They argued that, under conditions of experimental restraint, the mere repeated elicitation of aversive excitatory and inhibitory reflexes was enough to result in the development of experimental neurosis. Further, they found that simply restraining a previously trained sheep in the experimental harness (an apparatus they refer to as a "psychical strait jacket") for an hour session without any stimulus presentation whatsoever was sufficient to evoke evidence of pronounced autonomic and behavioral distress. Loss of control (i.e., exposure to uncontrollable and inescapable events) plays an integral role in the experiments of Liddell, which support the view that control over vital events is of critical importance for the elaboration and maintenance of adaptive behavior and vice versa. Mineka and Kihlstrom discuss the role of control and predictability in Liddell's experiments at length,

Initially Liddell's group attributed experimental neurosis to a variety of neuroendocrine changes. Later, however, Liddell offered an interpretation very much in accord with our own, noting that the domesticated animal already lives under conditions of considerable re-straint—a condition that is exacerbated by the exigencies of the laboratory experiment. Restraint, or loss of control, in either situation alone may be disturbing to the animal, and the two in conjunction are likely to be even more stressful. Liddell often emphasized that laboratory experiments which did not involve restraint of movement (as did the Pavlovian harness), such as insoluble mazes, did not produce experimental neurosis. (1978:265)

Liddell noted that many of the animals he tested exhibited vigorous efforts to break free when they were first placed into the experimental harness. These initial efforts to escape were regularly followed by a sudden lapse into a tense state of resignation. Younger animals accepted exposure to loss of control more readily than older ones. In older animals, the freedom reflex appeared to be more persistent and unyielding than in the more compliant and flexible younger ones. Liddell described a 1-year-old billy goat that unremittingly continued to struggle and butt while attached to the training harness, making him useless for experimental purposes. Similarly, Pavlov reported a case involving an outgoing and friendly dog that strongly rebelled against restraint in the harness. However, he gradually overcame the dog's reactivity by reciprocally inhibiting the freedom reflex with the elicitation of another more salient reflex incompatible with continued struggling—eating. Pavlov strongly emphasized the adaptive importance of the freedom reflex:

It is clear that the freedom reflex is one of the important reflexes, or, if we use a more general term, reactions, of living beings. This reflex has even yet to find its final recognition. In James' writings it is not even enumerated among the special human "instincts." But it is clear that if the animal were not provided with a reflex of protest against boundaries set to its freedom, the smallest obstacle in its path would interfere with the proper fulfillment of its natural functions. (1927/1960:12)

The repeated and inescapable stimulation of escape was considered by Liddell to be the causal locus of neurotic elaboration in his experiments. This viewpoint is consistent with more contemporary theories in which loss of control over significant biological events is considered instrumental in the generation of various behavioral disturbances, like learned helplessness and post-traumatic stress disorder, for example. The animals in Liddell's experiments appear to give up psychologically and thereby predispose themselves to the development of neurotic symptoms and their elaboration under the influence of repeated stimulation of aversive emotional reactions and the chronic habituation of the orienting response. Under conditions of fearful or aver-sive stimulation where an animal's control over the situation is impeded, and the orienting response habituated through persistent and monotonous elicitation, a growing sense of insecurity and heightened distress occur together with increased arousal and various dysfunctional efforts to adjust. Some of the more pronounced symptoms that he observed in experimentally distressed animals included hyperirritability, restlessness, insomnia, bizarre postural compulsions, dysfunctional reflex movements, eliminatory disturbances, and a tendency toward self-isolation when with conspecifics.

A study of reactive hypertension in dogs provides some suggestive evidence linking the chronic inhibition of the freedom reflex with increased stress and irritability. Wilhelm and colleagues (1953) found that highly trained and conditioned dogs exhibit a much more extreme blood pressure response to trivial stimulation (entry of a stranger, strange noise in the kennel, and change in routine) than less-well-trained counterparts. The authors speculated along with Liddell that repetitive and monotonous inhibitory training and conditioning is itself stressful and neuroto-genic. In a study assessing the effects of stress occurring during guide-dog training, stress-prone and non-stress-prone guide dogs were evaluated (Vincent and Mitchell, 1996). The researchers found that stress-prone dogs tended to exhibit significantly higher blood pressure readings when compared with non-stress-prone dogs, suggesting that temperament factors play an important role in the physiological expression of stress. These observations raise many questions with regard to training methodology and the various possible side effects resulting from excessive and boring inhibitory conditioning. In addition, the adverse effects of stressful training appear to depend on a dog's temperament and the dog's inclination toward reactive autonomic arousal.

One of the experiments performed by Liddell (1964) generated unusual and dramatic results. The objective of the study was to compare the effect of maternal contact on neurosis-inducing conditioning in twin goats. Siblings at 3 weeks of age were exposed to repeated leg flexion training in the presence of a dimmed-light CS: every 2 minutes, the light in the experimental room was dimmed for 10 seconds, followed by the presentation of mild shock. Twenty such trials were carried out daily over the course of training. The twins were stimulated identically, except one was kept with its mother during testing while the other was left alone.

Several remarkable outcomes resulted from the foregoing procedure. The twin conditioned in the presence of its mother learned to flex its leg as a kind of trick and did not develop the collateral emotional and behavioral disturbances associated with experimental neurosis. In contrast, the yoked twin, undergoing simultaneous conditioning in a separate room, became more reactive and upset as conditioning proceeded. Eventually, the isolated twin became immobile as though restrained in an invisible Pavlovian harness and subsequently exhibited many symptoms of experimental neurosis.

The foregoing experiment is an interesting confirmation of Gantt's effect-of-person theory. The twin conditioned in the presence of its mother was somehow protected from the ill-effects of the training procedure. Liddell observed an unexpected outcome of this series of experiments: all of the isolated twins died within a year (many within a few months) of various diseases, while the twins that were trained in the presence of their mothers survived into adulthood.

Experiments involving dogs performed by Liddell's group were carried out but not very often, as indicated by their lack in the literature. James (1943) utilized a conditioned avoidance response involving strong shock delivered to the foreleg of a German shepherd. The dog's leg had been weighted down so that, to avoid shock, the dog had to lift 30 pounds. The result was "total flight and escape behavior ... signaled in the kennel by the entrance of the experimenter rather than a specific signal in the laboratory (1943:117). After several months of rest, the first CS presented to the dog resulted in such a massive panic reaction that he broke out of the harness restraining him (Broadhurst, 1961). Cook (1939) reported a study by Drabovitch and Weger (1937, in French), who used a somewhat similar method as that described by Liddell. The dogs were trained to flex the left hind leg in response to a bell that was followed by shock. Once the flexion action occurred under the signalization of the bell alone, the experiment was modified in a way reminiscent of Yerofeyeva's experiment discussed earlier. Now, instead of attaching the electrode to the left hind leg, it was attached to the front left leg. The bell was rung repeat edly but without the delivery of shock. After three daily experimental sessions involving this arrangement, one of the dogs became quite agitated and on day 4 exhibited continuous convulsions of the left hind leg throughout the testing period.

A second dog involved in their study that had been exposed to leg flexion training over the course of 2 years was given a month-long break. During this period of rest, the dog's cage mate was removed and taken to the laboratory at the normal times he had been tested in the past. When the resting dog was tested after a month, he responded with unexpected and intense withdrawal reactions and convulsions of the left hind foot, a reaction that gradually generalized to the right hind leg as well. The combined convulsions resulted in the dog not being able to stand. Gradually, the dog became reactive as soon as he entered the laboratory situation, exhibiting a high degree of generalized arousal and withdrawal efforts. The researchers speculated that the daily excitement and distress of losing his cage mate may have adversely affected the dog's response to testing by "supercharging" centers controlling conditioned flexion. Although this explanation is not very appealing, the experiment does emphasize the potential inimical effects of separation distress on the elaboration of behavioral disturbances.

Dog Care Duty

Dog Care Duty

Get All The Support And Guidance You Need To Be A Success At taking Care Of Your Dog. This Book Is One Of The Most Valuable Resources In The World When It Comes To What Is Needed To Take Care Of Your Best Friend.

Get My Free Ebook

Post a comment