Classical Conditioning and Fear
Voluntary Versus Involuntary Behavior
Behavior can be roughly divided into two broad categories: voluntary (goal directed) and involuntary (reflexive). This division is not arbitrary but is based on the two fundamental ways behavior is modified. Voluntary behavior is highly goal directed and influenced by the consequences it produces. involuntary behavior, on the other hand, is largely composed of automatic mechanisms operating outside of a dog's volition and ability to choose. in the case of involuntary behavior, the presentation of a sufficiently salient stimulus evokes the elicitation of a highly predictable response. involuntary behavior is usually associated with simple reflexive and emotional responses. Although functionally independent of voluntary control, involuntary behavior is affected by antagonistic motivational states and modified through associative learning procedures. Together voluntary behavior and involuntary behavior provide an adaptive interface between a dog's changing biobehavioral needs and the surrounding environment.
Behavioral disorders are complex and problematical, consisting of both instrumental and reflexive (respondent) components. Defensive behavior involves two motivational processes, one under relatively more voluntary control (freeze, flight, and fight) and the other under relatively more involuntary or autonomic control (fear and anger). For example, individuals fearful of snakes cannot by an act of will persuade themselves not to feel afraid when confronted with a snake, but can, despite great apprehension and reluc tance, possess enough self-control not to run away. Consequently, to attenuate fearful behavior properly, one must address both instrumental fearful responses as well as the underlying emotional concomitants. From the perspective of some forms of behavior therapy, fear is best reduced by simply strengthening instrumental behavior incompatible with fear while ignoring or blocking (response prevention) fearful behavior when it happens to occur. Sometimes, however, the underlying fear is so strong and pervasive that it must first be addressed and modified through direct means, including respondent counterconditioning, relaxation training, exercise, or medications. As the underlying fearful arousal is diminished, the instrumental behavioral expressions of it will spontaneously improve, thus making it easier to shape more confident behavior.
Three Boys and a Brief History of Fear
Among the earliest controlled studies on the development of conditioned fear were those carried out by the American psychologist John Watson (1924/1970) and coworkers. Watson, often called the father of behaviorism, successfully conditioned a stable fear response in an 11-month-old orphan infant named Albert (Watson and Rayner, 1920). Albert was exposed to a white rat and observed for his reactions. Prior to conditioning, he was accustomed to holding and playing with the animal and exhibited no signs of fear. The fear-conditioning stimulus used by Watson was a startling sound made by striking a hammer against a heavy steel bar held behind the infant's head. As the child reached for the rat, the bar was struck, causing Albert to recoil from the animal. Over the course of several similar trials, Albert's fear deepened and became progressively more reactive and generalized. It was found during subsequent tests that Albert's fear had generalized to include other furred animals (a rabbit and a dog) and inanimate objects such as a fur coat. Although there were plans to "uncondition" Albert, he was subsequently adopted by a family living out of town. No one knows what finally became of Little Albert.
Additional work was carried out by Mary Cover Jones (1924) in an effort to isolate the most effective training techniques for reducing fearful behavior in children. Jones studied several methods, but the one she called direct unconditioning is of particular interest to dog trainers and behavior consultants. Peter, a 3-year-old boy, already exhibited intense fears toward various animals and furry objects. After trying several methods with varying degrees of success, the researchers exposed Peter to an early prototype of graded countercondi-tioning that turned out to be extremely effective. The method involved feeding Peter in the presence of the feared animal (a rabbit). The rabbit was initially caged and then systematically presented to the child at closer distances. The progress of these graded exposures was regulated by Peter's willingness to eat, based on an observation that relative appetite was a sensitive indicator of fear. The rabbit was gradually moved closer to the child through progressive steps, until finally he was able to eat without signs of anxiety while at the same time petting the animal who had been placed on his lap.
A decisive shift in the study of fear occurred with the publication of an article written by Wolpe and Rachman (1960). In the article, they criticize the psychoanalytical perspective on phobias and their development, especially with respect to Freud's interpretation of one of his cases involving a child named Little Hans. Little Hans, who was a patient of Freud, had acquired a strong phobia of horses. In a lengthy report, Freud concluded that the boy's fear was fueled by an underlying Oedipal conflict with his father. He argued that the horse was symbolically linked in the boy's mind with his father—the true object of the boy's fear. As the result of the child's forbidden wish to possess his mother sexually coupled with an unconscious desire to kill his father, the boy experienced a profound sense of guilt and transferred his fear of retribution to the horse. Wolpe and Rachman argued that little evidence in Freud's case study actually supported an Oedipal interpretation of the boy's fear. In reading the material, they discovered that the boy had witnessed a tumultuous accident in which a horse had fallen in the street while pulling a bus. The experience was a traumatic one for Hans and, the researchers argued, one sufficient to explain the boy's subsequent fear of horses.
Phobic Cats and Systematic Desensitization
Wolpe's (1958) experimental work with conditioned fear was carried out with cats from 1947 to 1948. Several cats were exposed to conditioning procedures that resulted in neurotic fear reactions to a variety of stimuli and situations. One group was conditioned to respond to the "hoot" of a horn that was followed by shock delivered into the floor grid of the experimental cage. The second group of cats was first trained to go to a dispenser of food in response to a buzzer signal. Once this behavior was well established, shock was delivered just before the cat took the available food into its mouth. Both groups developed strong phobic reactions toward the conditioned stimuli. The latter method is very similar in effect to that used by Watson on Little Albert. Unlike Albert, however, who showed no signs of generalized fear involving the experimental setting, Wolpe's cats not only exhibited intense fear toward the eliciting conditioned stimuli, they also resisted entering the cage, exhibited various signs of fear while in the cage, and refused to eat even after several days of continuous food deprivation while in the cage where the shock too place.
Wolpe studied several methods for extinguishing fear in cats. Some of the fearful cats were encouraged to eat food by pushing it toward them with a stick. Wolpe theorized that the cats would see the hand as a conditioned appetitive stimulus, since they had been previously fed by hand. A few of the cats did, in fact, respond to this method. Cats who failed to respond to this first method were exposed to a directive training procedure in which they were forced into close contact with an appetizing food item. Under these conditions, several additional cats were eventually coaxed into eating. Since many of the cats not only exhibited fear in the experimental cage but also to the surrounding room, Wolpe attempted to feed the remaining cats in four separate rooms of increasing similarity to the one in which the shock took place. This method proved very effective, allowing the cats eventually to eat readily from within the experimental cage itself.
The next training problem was to devise procedures for reducing the cats' fearfulness and avoidance in the presence of the buzzer or horn sounds. This training goal was accomplished along two lines of conditioning. The first method was very similar to that used by Jones with the phobic child Peter. Wolpe determined that cats would remain relatively undisturbed if the hooter was sounded at a distance of at least 40 feet away. This distance could be progressively decreased by gradual successive approximations. This method was carried out until the cats accepted the sound of the horn and buzzer at full conditioning intensity. An alternative method he employed to attenuate the intensity of the eliciting CS was accomplished by truncating its duration to a fraction of a second. The first sample was presented at full volume for a fifth of a second. Under such conditions, the cats would initially react and, after a short delay (about 40 seconds), would consume the food pellet dropped into the cage. Gradually, over several trials, the delay between the CS sample and eating decreased until at last the cats were taking the food within a couple of seconds following the CS. The duration of the fear-eliciting CS was then gradually increased until the cats would continue eating under a continuous presentation of the CS for 30 seconds.
Reciprocal Inhibition
Wolpe attributed the effectiveness of his training efforts to reciprocal inhibition. Essentially, reciprocal inhibition postulates a hypothetical interference occurring when two opposing emotional states are simultaneously elicited. Two hedonically opposing emotional states cannot exist simultaneously: one overshadows or offsets the other. in the case of Wolpe's fearful cats, they could not be interested in food while at the same time feeling anxious; appetite in this case overshadows or reciprocally inhibits fear. This process of countering aversive emotional arousal by elic iting stronger appetitive arousal or relaxation in the presence of the aversive event is referred to as counterconditioning. Wolpe states the underlying principles facilitating the effects of counterconditioning as follows:
If a response inhibitory of anxiety can be made to occur in the presence of anxiety-evoking stimuli it will weaken the bond between these stimuli and anxiety (1969:14).
Negative emotional reactions evoked by fear-eliciting conditioned stimuli can be systematically reduced, modified, or replaced with more adaptive and positive response patterns through counterconditioning. Many undesirable conditioned emotional reactions are traceable to some past event and are learned in a manner consistent with the experimental models devised by Watson and Wolpe. The goal of counterconditioning is to disassociate past learning from the eliciting CS and to establish new, more positive associations controlled by the same stimulus.
Graded Counterconditioning
Fearful emotional responses are subdued by the elicitation of competing incompatible emotional responses (e.g., appetite or relaxation). The central maxim informing this process is "contraries are cured by contraries," that is, two hedonically opposed emotional responses cannot exist at the same time—the elicitation of one reciprocally inhibits the other. For example, a strange noise occurring while a dog is happily chewing on a fresh bone is less likely to elicit fear than is a similar stimulus occurring at another time when the dog is not so preoccupied. In this case, the appetitive interest evoked by the bone reciprocally inhibits fear elicited by the strange noise. Other important factors affecting counterconditioning, as both Wolpe and Jones have demonstrated, is the intensity and proximity of the fear-eliciting CS. If the fear-eliciting CS is too strong or close, the incompatible counterconditioning stimulus may be overshadowed and the process impaired. For example, a xenophobic dog may not notice a stranger walking 100 feet away but will react with intense fear if the same person approaches too closely or attempts to make physical contact. For effective countercondi-tioning to occur, the dog must be gradually exposed to strangers at progressively closer distances and under increasing levels of provocativeness while the dog is concurrently stimulated by a strong counterconditioning stimulus.
The best counterconditioning results are achieved by presenting stimuli that either relax a dog or satisfy it appetitively while systematically exposing it to the fear-eliciting target. Relaxation and eating are incompatible with fear—that is, a dog cannot be simultaneously fearful while relaxing or eating. Some activities like playing, running, and even walking can be used as countercondi-tioning stimuli to reduce mild fears and anxieties. For convenience, food is usually chosen as the primary counterconditioning stimulus, although massage can be used effectively in some situations. The course of systematic de-sensitization follows a regular pattern. A hungry dog is progressively exposed to the feared object through a series of defined steps (a hierarchy of fear-eliciting stimuli), which enables closer proximity and, finally, direct contact without eliciting fear at any point. Each step of this hierarchy is associated with food and reassurance, providing a secure foundation for the next step. The dog learns to associate good things with the feared object, gradually abandoning its fearful attitude in exchange for a more positive expectation.
Conditioned fear is frequently very resistant to normal extinction procedures. Since fears and anxiety may not be attenuated under normal conditions, special methods must be employed to achieve the desired effect. In addition to counterconditioning, a key element in the reduction of anxiety and fear is controlled exposure that allows the dog to engage in direct interaction with the feared object/situation.
Interactive Exposure and Flooding
The reduction of fearful behavior is facilitated by utilizing a combination of behavioral training methods. In addition to graded counterconditioning, several other fear-reducing techniques have proven efficacious in reducing fear. Jersild and Holmes (1935) pro vide an important historical study detailing the broad aspects of graded interactive expo-sure—a method that has proven complementary to Wolpe's model and, according to some authorities, should replace it as the treatment strategy of first choice (Marks, 1977). Jersild and Holmes observed that two primary methods of confronting fear are most commonly used by parents of fearful children: direct repeated exposure to the feared object and ridicule. Of the two methods, direct exposure appears to be the most effective, resulting in the reduction of fear in 50% of the cases studied. Ridicule and invidious comments (e.g., wimp, "scar'ty cat") yield no benefit in bolstering a child's courage. They found that common childhood fears could be systematically reduced by replacing avoidant behavior with confidence-building interactive skills, resulting in the progressive development of competence in the fearful situation. Their method emphasizes developing various coping skills and participatory activities that the child engages in while in direct contact with the feared object/situation. They utilized attractive coun-terconditioning stimuli, not intended to change associative responses directly but to provide additional incentives to the child for making such contact. The countercondition-ing stimulus serves as a bait to lure the child into sustained interactive contact with the feared object/situation. Similar benefits may be derived in the case of moderately fearful dogs. in many particulars, the Jersild and Holmes method anticipates current methods for reducing fear in dogs.
Since Wolpe's discovery, many studies have been carried out to evaluate the therapeutic efficacy of the desensitization and the counterconditioning process. The results have often been critical of Wolpe's conceptualization on several grounds. For instance, the need for a hierarchy of fear-evoking conditioned stimuli presented systematically in the presence of hedonically antagonistic counter-conditioning stimuli (eating or relaxation) has proven relatively unimportant under experimentally controlled conditions (Delprato, 1973; Delprato and Jackson, 1973). In fact, according to Delprato's study (1973), simple extinction proved more effective than both systematic desensitization (graded counter-conditioning) and graded exposure. Since the desensitization analogue used by Delprato differed from graded exposure only in terms of the presentation of food, it might even be further argued that the presentation of food actually interferes with the reduction of fear. One possible explanation he presented to explain this unexpected result is that the acquisition of food during graded exposure overshadows an animal's attention to the fear-eliciting stimulus. Unlike in the cases of graded exposure and simple extinction, the animal may have failed to learn that the CS no longer predicts a pending aversive event. Subsequently, when exposed to the fear-eliciting stimulus without food, the countercondi-tioned animal exhibited little or no improvement in comparison to controls.
The most important factor in the desensi-tization process appears to be sustained exposure to the fear-evoking stimulus until fear subsides (Marks, 1977). This procedure is commonly referred to as flooding through response prevention. Response prevention and direct exposure (flooding) can be carried out in the presence of full-intensity samples of the fear-eliciting stimulus or, more optimally, a progression of increasingly intense samples. A precaution needs to be carefully observed: if dogs are fearful when the flooding exposure is terminated, their fearfulness might be made worse. Also, there is some evidence that frequent brief exposures to the feared stimulus situation might actually strengthen the reaction rather than weaken it. Therefore, it is important that fearful exposures be of sufficient duration at each step to allow elicited fears to habituate before proceeding to the next step or before quitting.
Response Prevention and Directive Training
Whether counterconditioning or interactive exposure is used, additional supportive techniques may be required that force a dog into direct contact with the feared CS complex. At some point in the process, the dog may become overly reactive or attempt to escape from the situation. While care should be taken not to overwhelm fearful dogs with ex posures that they are unable to tolerate, when avoidance responding does appear it must be blocked or corrected (Askew, 1997). In some situations, dogs may completely refuse to engage in some behavior as the result of competing avoidance responding, such as climbing stair steps or entering certain rooms or places. In such cases, directive exposure involving the use of a series of corrective prompts on the leash is often both expedient and very effective.
Various response prevention techniques can be selectively applied during countercon-ditioning and flooding efforts. Occasionally, dogs undergoing rehabilitative training will persist in phobic avoidance behavior despite gradual and patient efforts. One theory suggests that avoidance responding actually forestalls the unlearning of fearful responses (Levis, 1979). Persistent avoidance responding prevents dogs from coming into direct contact with the fear-evoking stimulus, thereby impeding the normal extinction process. Experiments designed to block or prevent avoidance responding have shown that extinction of such behavior is facilitated by such measures (Baum, 1970). Response prevention involves physically restraining a dog so that the avoidance response cannot be performed, requiring instead that the dog directly experience the avoidance-signaling CS while being prompted to perform some incompatible response (e.g., sitting or lying down) that is consequently rewarded. Such training provides the dog with coping options that supersede the avoidance response and may eventually take its place. The foregoing discussion of fear and its management anticipates a more thorough treatment of the subject in Volume 2.
Post a comment