Reinstatement of fear is defined as the reappearance of fear that has undergone partial or complete extinction. This return may be influenced by factors that were present before, after, and during treatment. Episodes of the return of anxiety often occur along with stressful events or major changes in a person’s life. This summarized chapter focuses on exposure and the return of anxiety. In this context, use if made of theory and research on extinction with classical conditioning. In addition, human conditioning is discussed. Thirdly, two posttreatment manipulations are examined: renewal (changing the context after extinction) and reinstatement (the presentation of unpredicted unconditioned stimuli (USs) after extinction).

Exposure and Extinction

The behavioural treatment of anxiety disorders often includes systematic and repeated exposure to the stimulus that evokes anxiety. Exposure corresponds to extinction on some points. During extinction, the conditioned stimulus (CS) is repeatedly presented without the US. During exposure, the patient is repeatedly exposed to the stimulus that generates anxiety. However, extinction is not the same as un-learning, but rather involves an additional learning experience. Two associations arise through extinction: a CS-US association/excitatory link and a CS-noUS association/inhibitory link. Four postextinction restoration phenomena prove that extinction is not the unlearning of an association: spontaneous recovery (return after the mere passage of time), disinhibition (return after the presentation of a novel stimulus), reinstatement (return after re-exposure to the US), and renewal (return after a context change). The best explanation is that exposure treatment temporarily or contextually masks the fear-arousing association but does not erase it.

Underlying Mechanisms of Extinction

The majority of knowledge about underlying mechanisms of extinction come from animal studies though there are differences between conditioning research in humans and animals. In animal conditioning, acquisition and extinction are under full experimental control. In human studies, there is no control over the acquisition of fear/anxiety. A solution to this lack of control is a laboratory experiment in which fear is established before it is eliminated. the strength of conditioning studies is that they distill theoretical assumptions to their essence and allow studies of the phenomenon under strictly controlled circumstances, hereby permitting better analysis of the underlying mechanisms. Of the different human conditioning paradigms, the human fear-conditioning paradigm is the paradigm that comes closest to a real-life fear situation.

Human Fear Conditioning

In the conditioning paradigm, neutral visual stimuli are used as CS and either an electrocutaneous stimulus or a loud aversive noise is used as US. One of the two visual stimuli is systematically followed by a US (the CS+ and the other not (the CS-). With the extinction, both stimuli are presented without the US. Associative learning effects are measured by a decrease or increase in differential responses to the CS+ and CS-. Because emotions can only be measured through behavioural, verbal, or psychosocial changes indexes for this have also been included in the study. The physiological indexes are skin conduction and shock modulation. The behavioural indexes are affective priming and the secondary reaction time task. Affective priming uses positive and negative words for the CS_ or CS- to indirectly measure valence. The secondary response time task emits a tone during the presentation of CS+ and CS-. The participant must press a button when he hears the tone. Response times are expected to be slower with the CS+ than with the CS-. On the basis of thee evidence, it’s concluded that CS+ after acquisition has become a fearful stimulus, as defined by properties of negative valence: high arousal.

The data related to extinction results from this experiments are: conditioned verbal expectancy from the US, electrodermal responding, shock responses in the form of blinking eyes, and the response times from the secondary response time task. In the paradigm used here, it’s found that that the CS+ becomes a fearful stimulus to acquisition and at the end of a standard extinction procedure, the CS+ has lost at least one essential aspect of fear, namely, US expectancy. However, the valence of the CS+ seems more difficult to change.

Renewal of Fear

Renewal is defined as the return of extinguished conditioned responses caused by changes in the contextual cues that were present during extinction. The most observed form is ABA (being tested in the original environment, while extinction took place in a different environment). ABC and AAB are also found.

Evidence from the clinical literature also emphasizes the importance of the environment. Acquisition easily generalizes to a new context, but extinction does not. Within the paradigm used, evidence has been found that extinction in a context other than that in which the acquisition took place or with a stimulus other than the acquired stimulus can cause the return of fear. The aforementioned difference in generalization between acquisition and extinction is not found in an AAB design, whereby the acquisition and extinction took place in the same context. Renewal in AAB appears to be more difficult to obtain than in an ABA design. That AAB effects are weaker than ABA effects appears to be useful for therapy: it can help to imitate the original acquisition situation during therapy (if possible).

Mechanisms of Renewal and Clinical Implications

Bouton and his colleagues developed a contextual theory about extinction: the context of extinction has a modulating role and helps to eliminate ambiguity between old knowledge (CS-US association and acquisition) and the new knowledge (CS-no US association or extinction). In this theory, renewal is a consequence of leaving the extinction context. Two alternatives are suggested to explain the context specificity of extinction:

• Forming a direct inhibiting association between the context of extinction and the US. The patient learns in practice that the context of therapy is safe.
• Perceiving the fearful object as a partially different stimulus in the context of exposure than in the context of acquisition. In the literature on animal conditioning, this is called generalization decrement.

Prevention of Renewal

In the literature on animal conditioning, two strategies are proposed to stimulate transfer to other contexts and prevent renewal. The first is the use of retrieval cues: stimuli that can be taken from the context of extinction to the test context and retrieve information from the extinction-exposure episode. The second method is extinction in multiple contexts. In humans, the effectiveness of retrieval cues has been demonstrated by means of a differential human contingent learning experiment. It was also found that manipulating contexts effectively reduces the renewal of fear of spiders. The method that uses retrieval cues does not change what has been learned during exposure: the association still depends on the context. The method that uses multiple contexts, on the other hand, effectively increases the number of contexts that may have an impact on what has been learned. Two mechanisms are important to explain renewal: the modulating mechanisms (context that modulates the CS-US relationship) and the direct inhibiting association between the context and the US.

Reinstatement of Fear

Reinstatement is defined as the return of extinguished conditioned responses caused by the experience of one or more US-only presentations after extinction. An experiment to stimulate this first uses the acquisition of a conditioned response through the contingent presentation of a CS and a US. Subsequently, the CS is repeatedly presented alone, causing extinction. The US is then shown without the CS being displayed.

Reinstatement appears to be a robust phenomenon. But research shows that conditioned contextual stimuli are necessary to obtain reinstatement. To deal with this context-dependency observation, the context-CS summation view was formulated. During the reinstatement phase, reinstatement context becomes excitatory, and this contextual conditioning is assumed to summate with the residual associative strength of the CS after extinction.

Another explanation is that reinstatement is seen as a special event caused by contextual cues. Leaving the context of extinction makes it more difficult to retrieve the CS-noUS association.

Mechanisms of Reinstatement and Clinical Implications

The fear-conditioning paradigm allows for the manipulation of several variables that may elucidate the mechanisms that drive reinstatement. For example, if people also find that reinstatement can take place with a US differing from the original US, there are many more possible circumstances in which fear can arise again. In addition, it is true that reinstatement can, to a certain extent, be considered a special case of renewal, this would suggest that methods available to reduce renewal may be applicable to attenuate reinstatement.

For human conditioning, it remains unclear whether reinstatement requires the ‘new’ US to share properties with the original. As described earlier, the valence of a stimulus doesn’t seem to disappear due to extinction. The authors of this chapter suggest that valence is an explanation for individual difference in the strength of the reinstatement effect.