Principles of Treatment

The treatment model that derives from classical conditioning states that behaviors and emotions can be changed by disrupting the associations that have formed between a cue (CS) and either an aversive or a pleasant outcome (US). In learning theory, this is referred to as extinction. Conditioning involves pairings of the CS with the US; extinction involves repeated presentations of the CS without the US. The corresponding treatment is referred to as exposure therapy; in this therapy, the client repeatedly faces the object of fear (in the case of anxiety disorders) or the drug-related cue (in the case of substance use disorders) in the absence of an aversive or a pleasant outcome. As an example, individuals with social anxiety would be encouraged to repeatedly enter social situations without being ridiculed or rejected, or individuals with posttraumatic stress disorder would be encouraged to repeatedly enter places where they were previously traumatized without being retraumatized. As another example, individuals who drink alcohol excessively would be exposed to substance cues (e.g., sight or smell of alcohol) and prevented from consuming the alcohol so that the CS is repeatedly presented in the absence of reinforcement that comes from the consumption of the drug. This is called cue exposure.

Several mechanisms are believed to underlie extinction and thereby exposure therapy. One such mechanism is habituation (or decreased response strength simply as a function of repeated exposure). Another mechanism, inhibitory learning, is considered to be even more central to extinction (Myers & Davis, 2007). Inhibitory learning means that the original association between a CS and aversive event is not erased throughout extinction, but rather a new inhibitory association (or expectancy) is developed. For example, as a result of exposure therapy for fear of dogs, an original “excitatory” association between a dog and ferocious barking would be complemented by a new “inhibitory” association between a dog and the absence of ferocious barking. Consequently, as a result of exposure therapy, two sets of associations exist in memory. Once exposure therapy is over, the level of fear that is expressed when a dog is encountered in daily life will depend on which set of associations is evoked. Interestingly, basic research by Bouton and colleagues (reviewed in Bouton, Woods, Moody, Sunsay, & Garcia-Gutierrez, 2006) indicates that context is important in determining which set of associations is evoked. If the previously feared stimulus is encountered in a context that is similar to the extinction/exposure therapy context, then the inhibitory association will be more likely to be activated, resulting in minimal fear. However, if the previously feared stimulus is encountered in a context distinctly different from the extinction/exposure therapy context, then the original excitatory association is more likely to be activated, resulting in more fear. Following the example of dog phobia, assume that the exposure treatment was conducted in a dog training center. Then, once treatment is over, a dog is encountered on a neighborhood sidewalk, a context that is distinctly different from the dog training center. On the sidewalk, the original excitatory fear association is more likely to be activated than the new inhibitory association that was developed throughout exposure treatment, resulting in the expression of fear.

Thus, a change in context is presumed to at least partially account for the return of fear that sometimes occurs following exposure therapy for anxiety disorders (Craske et al., 2008) and relapse following treatment for substance use disorders (e.g., Collins & Brandon, 2002). In addition to context, other factors can also reactivate the original excitatory association. One such factor is being exposed to a new negative experience. Thus, persons who are successfully treated for their fear of dogs may have their fear return if they are subsequently involved in a car accident (in learning theory this is called reinstatement) or if they are barked at by another ferocious dog (termed reacquisition).

Innovative strategies are now being tested for enhancing new inhibitory associations throughout exposure therapy (see Craske et al., 2008, for a review). In addition, attention is being given to ways of enhancing the retrievability of new inhibitory associations once exposure therapy is completed, and thereby decreasing relapse, such as conducting exposure therapy in multiple contexts. Another is to provide retrieval cues that remind clients, when they are outside of the therapy context, of the new learning that took place in the therapy context or at least recommend to clients that they actively try to remember what they learned when in the therapy context (see Craske et al., 2008).

Another key concept associated with extinction of CRs is safety signals, or conditional inhibitors that predict the absence of the aversive stimulus. When the conditional inhibitor is present, the CS is not paired with the US; when the conditional inhibitor is not present, the CS is paired with the US. In the experimental literature, safety signals alleviate distress to the CS in the short term, but when no longer present, fear to the CS returns (Lovibond, Davis, & O’Flaherty, 2000). Common safety signals for anxiety disorder clients are the presence of another person, therapists, medications, food, or drink. Thus, clients with panic disorder and agoraphobia may feel relatively comfortable walking around a shopping mall with a bottle of medications in their pocket (even if the medications are never taken) but report being anxious in the shopping mall when without the bottle of medication. Conditional inhibitors have been shown to interfere with extinction learning in human experimental studies (e.g., Lovibond et al., 2000).

Also, several studies have evaluated safety signals in phobic samples (see Craske et al., 2008, for a review). For example, claustrophobic participants who are encouraged to use safety signals during exposure to being inside a small booth report more fear at later testing to the claustrophobic booth without the safety signals than those who complete exposure without safety signals. (The safety signals were to open a window on the side of the booth and to check that the booth door unlocked.) Just the perception of safety (i.e., knowing that the safety signals were available even though they were not used) has the same detrimental effects on outcome as the actual use of safety signals. Thus, exposure therapy typically proceeds by not only having clients repeatedly face their feared objects or situations, but at the same time being weaned from typical safety signals.