Stimulus Factors Affecting Conditioned Stimulus Acquisition and Maintenance

External Inhibition and Disinhibition

Even after a CS has been well established, it may undergo further potentiation or attenuation under the influence of various internal and external events impinging on the central mechanisms controlling it. Both excitatory and inhibitory conditioned stimuli are subject to such change. Dramatic examples of external inhibition and disinhibition can be observed among dogs fearful of loud noises or subject to separation distress when left alone. During thunderstorms or fireworks, such dogs are often overcome with fear and may lose control of many previously well-conditioned habits. A startling noise may cause otherwise well-trained dogs to pull frantically out of harm's way, even though danger never actually threatened them. Dogs reactive to separation may lose control of bladder and bowel functions, howl and bark continuously, or become destructive toward owner belongings. The effects of external inhibition or disinhibition can never be entirely eliminated. Well-trained dogs should be proofed against these influences through graduated exposure to diverse environments and by training them under progressively stressful conditions.

Conditioned Inhibition

Once a CS reliably predicts the occurrence of the US, it becomes an excitatory stimulus (CS+) for response properties controlled originally only by the US. Opposing inhibitory conditioning occurs when the CS is presented in the absence of the US. The inhibitory CS (CS ) predicts the nonoccur-rence of the US. For instance, if a dog is differentially exposed to a light that always precedes food and a tone that always precedes the omission of food, the light will become an excitatory stimulus (CS+) for food and the

Fig. 6.5. Relationship between expectancy and classical conditioning. CS, conditioned stimulus; US, unconditioned stimulus.

212 Chapter Six tone an inhibitory stimulus (CS predicting the absence of food. Later, if the experimenter decided to reverse this arrangement by making the tone predictive for food instead of signaling its omission, the dog would learn this contrary association much more slowly than if the stimulus were neutral. This impediment results from previously learned stimulus associations competing with current training efforts. Dogs must first unlearn what they have already learned about the tone (i.e., the tone must first be disconfirmed as a predictor of no food) before it can become an excitatory stimulus predicting the presentation of food.

Pavlov studied conditioned inhibition as a phenomenon occurring between excitatory and inhibitory conditioned stimuli. Conditioned inhibition occurs when a previously acquired excitatory CS is presented in combination with an inhibitory CS. Take, for example, a dog that has been trained to respond to a bell as a CS for food. On all occasions when the bell is presented alone, it is followed by food. Now consider what happens if the bell is intermittently presented in combination with a tone, but whenever the bell and tone are presented together, the food is omitted. Over time, the tone (CS-) will restrain the excitatory effects of the bell (CS+) when the two stimuli are presented together as a compound stimulus:

bell (CS+) > food::salivation bell (CS+) & tone (CS-) > no food::

reduced salivation

The dog has learned that the presentation of a compound stimulus composed of a bell and tone stimulus predicts the absence of food. If the inhibitory tone stimulus (CS ) is now combined with some other previously conditioned excitatory stimulus (CS+), for example, a light, it will be found that the inhibitory effect obtained by presenting the bell and tone together without food transfers to control this remote CS+ (light). When the light (CS+) is presented with the tone (CS-), salivation normally elicited by the light CS+ is inhibited (Fig. 6.6).

Further, as just noted, if the tone (a condi tioned inhibitor) is now paired with food to make it an excitatory CS, it is found that this is a rather more difficult process. It takes the dog longer to learn that the tone predicts food because this new association conflicts with a previously well-established contrary association—that is, the tone predicts the absence of food. This so-called retardation of acquisition effect can be observed in many training contexts. Dogs regularly exposed to CS events not followed by expected US events learn to treat such impinging signals and stimuli as irrelevant. Effective use of classical conditioning requires that dogs be exposed to clear and predictable occurrences of the CS preceding the US. Classical learning is never inactive: it provides inquisitive dogs with information regarding either the occurrence or nonoccurrence of important events—that is, the dogs are always learning to respond or not to respond.

Findings such as the foregoing suggest that both excitatory and inhibitory influences affect the CS. The excitatory CS (CS+) predicts the occurrence of the US, whereas the inhibitory CS (CS ) predicts the absence of the US (Fig. 6.7). These excitatory and inhibitory influences extend equally to attractive and aversive stimuli. Taken together, these various relations produce four types of classical conditioning (Fig. 6.8). Under natural conditions, the actual strength of the CS is a composite of CS+ and CS— influences, with the valence of the particular CS depending on the extent to which it predicts the presence or absence of the US.

Latent Inhibition

Repetitious presentation of a NS independently of the US results in the NS becoming associatively resistant to future classical conditioning. For example, if a dog's name is used casually (without evoking an appropriate attending response), the attention-controlling and orienting function of the name will be compromised and rendered more difficult to learn later on. Studies have shown that as few as 15 to 20 nonreinforced presentations of the NS prior to conditioning are sufficient to produce latent inhibition (Lubow, 1973). Animals exposed to such treatment fail to attend to the stimulus because its presentation has proven to be uneventful in the past, producing a cognitive interference effect that Baker (1976) refers to as learned irrelevance. If a dog has inadvertently learned that the CS is irrelevant, this interfering conviction must first be disconfirmed before new learning can take place. Classical learning appears to proceed most efficiently under circumstances where a completely novel CS occurs contiguously with a startling or surprising US.

Sensory Preconditioning

An interesting conditioning phenomenon occurs when neutral stimuli are paired together prior to conditioning (Fig. 6.9). For example, if the sound of a clicker occurs just prior to the word cue "Good" over several trials, an associative connection between these signals will occur even though the arrangement is not followed by a US. Evidence for the effectiveness of preconditioned associations becomes apparent only after the CS "Good" undergoes some actual conditioning with the US (e.g., food). Once such conditioning

100 Dog Training Tips

100 Dog Training Tips

100 Dog Training Tips EVERY Dog Owner Should Know. Utilize These Tips to Train Your Dog to Be Obedient and Smart.

Get My Free Ebook

Post a comment