provided the first report of extinction in his studies of digestive physiology in dogs. He observed that the conditioned salivation of dogs in response to an external food-signaling cue slowly decreased and eventually disappeared when the cue was presented repeatedly in the absence of food. The inhibition of fear acquired by associative learning has been studied in both animals and humans.48
Following the pairing of an aversive unconditioned stimulus to a neutral conditioned stimulus a conditioned fear response is established. If the neutral conditioned stimulus is then repeatedly presented in the absence of the unconditioned stimulus, a procedure known as extinction training, the result is the loss of the conditioned fear to the neutral conditioned stimulus. From an operational perspective, extinction may thus be defined as “a reduction in the strength or probability of a conditioned fear response as a consequence of repeated presentation of the conditioned stimulus in the absence of the unconditioned stimulus.”48
A variety of behavioral observations support the hypothesis that extinction is a form of learning and not “unlearning” or the forgetting of a conditioned association.49,50
The most nonspecific of these observations is the phenomenon of spontaneous recovery. This refers to the reappearance over time of a conditioned fear that had been previously extinguished through extinction training. An additional characteristic of extinction is context specificity, a phenomenon known as renewal. This means that extinction seems to be specific for the context in which the extinction training occurs and that the previously extinguished conditioned fear will return if tested in a new context. Finally, the phenomenon of reinstatement occurs when unsignaled presentations of the unconditioned stimulus or other stressor interrupt extinction and lead to re-emergence of the previously diminished conditioned fear response. Collectively, these data suggest that extinction is a labile form of learning that is specific with respect to environmental and temporal context and is vulnerable to degradation by stress and the passage of time.
Glutamate and the Neurobiology of Extinction
The neurobiology of extinction has not been studied to the same degree as that of fear learning. Data obtained with rodents51,52
indicate that extinction appears to be dependent on events occurring within, and interactions between, the prefrontal cortex and the amygdala. Functional brain imaging studies in humans53
are consistent with this hypothesis and demonstrate engagement of the amygdala and ventral medial prefrontal cortex during acquisition training and early extinction.
Several lines of data suggest that glutamate has a central role in this process. Virtually all of this work has focused on the role of the NMDA receptor in extinction and will thus form the focus for this portion of our review. Like associative fear conditioning, extinction is dependent upon activation of NMDA receptors and the induction of neuroplasticity to form new patterns of connectivity that facilitate alternative forms of behavior in response to a previously conditioned stimulus. Administration of NMDA receptor antagonists either systemically54,55
or by direct infusion into the basolateral complex of the amygdala25,56
prior to extinction training blocks the extinction of fear memories. In addition, other investigators have found that blockade of NMDA receptors after extinction training also impairs extinction suggesting that NMDA receptors participate in the consolidation of extinction memories.57
Finally, bilateral infusions of the mitogen-activated protein kinase inhibitor PD58089 into the basolateral complex of the amygdala impair extinction of previously acquired fear-potentiated startle,58
suggesting that the effects of extinction are dependent on the induction of neuroplasticity.
Extinction can be Enhanced with Partial N-methyl-d-aspartate Agonists
In contrast to previous experiments showing extinction to be dependent on the functional integrity of NMDA receptors or neuroplasticity pathways, Walker and colleagues59
tested the reciprocal hypothesis that enhancing neurotransmission at NMDA receptors would facilitate extinction. Because administration of full agonists at the NMDA receptor is associated with excitotoxic effects on neurons,60
the partial NMDA agonist D-cycloserine was used. D-cycloserine acts at the strychnine-insensitive glycine recognition site of the NMDA receptor complex to enhance NMDA receptor activity and has previously been used in humans for the treatment of tuberculosis. The central findings of this study were that systemic administration of D-cycloserine dose-dependently enhanced extinction of previously conditioned fear-potentiated startle but did not influence fear-potentiated startle in rats that had not received extinction training. Similar effects on extinction were found when D-cycloserine was given by infusion into the basolateral complex of the amygdala and appear to be pharmacologically specific as co-administration of the NMDA glycine recognition site antagonist, HA966, with D-cycloserine blocked the effects of D-cycloserine on extinction. The general findings of this study have been replicated by studies61-63
that have used a cue-conditioned freezing paradigm showing that extinction occur in a time-dependent manner suggesting an effect on consolidation. In addition, this same group has demonstrated that post-extinction training administration of D-cycloserine interferes with reinstatement of conditioned fear63
and recently reported that extinction training enhanced by D-cycloserine appears to result in generalized extinction such that fear behavior elicited by a nonextinguished conditioned stimulus is reduced.61
Collectively, the data from these rodent studies suggest that D-cycloserine, a drug already shown to be safe for use in humans, may have significant potential use in the facilitation of extinction-based therapies for human anxiety disorders.
Glutamatergic Therapies for the Treatment of Anxiety in Humans
Preclinical data regarding the properties of associative learning and extinction have enriched our understanding of the etiology of anxiety disorders and have been applied to the treatment of pathological anxiety in humans. From a therapeutic standpoint, the behavior therapies for different anxiety disorders generally involve some form of extinction training.48
This involves graded exposure to the feared object or event in the absence of any likely actual harm or where there is a small chance that harm may occur. This exposure may be imaginal in nature, wherein a narrative is read or listened to by the patient or in vivo where the feared stimulus is directly encountered by the patient. Additional variables are the within-session duration of the exposure and the number of sessions. Goals of this form of therapy are reductions in the frequency and intensity of the reflexive conditioned response (fear response) as the conditioned stimulus (fear stimulus) is repeatedly encountered without adverse effect.
Considering the similarity between extinction training in rodents and exposure therapy for anxiety disorders in humans, we were interested in finding novel ways to integrate pharmacotherapy with psychotherapy. Historically, there has been hope to combine these two approaches into a treatment more effective than either alone; unfortunately this has not been achieved.64,65
In fact, sometimes combining pharmacotherapy with psychotherapy can make a bad situation worse.66,67
However, extinction-based therapies for anxiety may be an exception to this trend. We believe that the preclinical literature on the neurobiology of extinction provides valuable insight into mechanisms, whereby behavior therapy approaches utilizing exposure protocols might be made more effective with adjunctive use of medication.
D-cycloserine Enhances Extinction of Fear of Heights in Patients with Acrophobia
Recently, Ressler and colleagues68
demonstrated that D-cycloserine facilitates psychotherapy for the treatment of specific phobia in humans. We wished to examine the ability of D-cycloserine to enhance emotional learning in humans using the most optimally controlled form of psychotherapeutic learning available. Virtual reality exposure (VRE) therapy is ideal for clinical research assessment because exposure and testing is identical between patients, is well controlled by the therapist, and occurs within the spatial and temporal confines of the limited therapy environment.69
This method has proven to be successful for the treatment of specific phobias and PTSD.69-71
With VRE for fear of heights we used a virtual glass elevator, in which participants stood while wearing a VRE helmet and were able to peer over a virtual railing. Previous research69
has shown improvements on all acrophobia outcome measures for treated as compared with untreated groups after 7 weekly therapy sessions.
To examine whether D-cycloserine would enhance the learning that occurs during exposure therapy for humans with specific phobia, Ressler and colleagues68
enrolled 28 volunteers who were diagnosed with clinical criteria acrophobia. Participants were randomly assigned to three treatment groups, placebo + VRE therapy, or D-cycloserine + VRE Therapy at two different doses of D-cycloserine (50 mg/day or 500 mg/day). Treatment condition was double-blinded, such that the subjects, therapists, and assessors were not aware of assigned study medication condition. Although Ressler and colleagues68
used two different doses of D-cycloserine, preliminary analysis of our data indicated that there were no significant differences between the 50 mg/day and 500 mg/day drug groups for the primary outcome measures of acrophobia. Therefore, Ressler and colleagues68
combined the two drug groups for analysis.
Participants underwent two therapy sessions, which is a suboptimal amount of exposure therapy for acrophobia.69
They were instructed to take a single pill of study medication 2–4 hours before each therapy session, such that only two pills were taken for the entire study. A posttreatment assessment was performed within 1 week following the two therapy sessions, and an additional follow-up assessment was performed 3 months after the therapy.
At 1–2 weeks and 3 months posttreatment, subjects that received D-cycloserine in conjunction with VRE therapy had significantly enhanced decreases in fear within the virtual environment (, P<.05). Furthermore, within the virtual environment, skin conductance fluctuations, a psychophysiological measure of anxiety, was significantly decreased in the group that received D-cycloserine in conjunction with therapy.
FIGURE 1 Acrophobia within the virtual environment is improved with D-Cycloserine68
One of the cardinal features of extinction in animal models is the context specificity of the extinction environment. However, Ledgerwood and colleagues61
have demonstrated that D-cycloserine enhancement of extinction in animal models appears to lead to generalization across contexts. Ressler and colleagues,68
therefore, wondered if the decreased fear of heights found within the virtual environment would generalize to other settings. This question was assessed in two ways: first by asking questions related to the subject’s fear of heights in the real world, and secondly by assessing how much the subjects had decreased their avoidance of heights since the treatment. We found that all measures of fear were decreased at the early assessment (not shown) and the 3 month assessment (). The researchers also found that subjects’ self-exposure to heights in the “real world” had increased, suggesting decreased avoidance (). Finally, subjects that received D-cycloserine in conjunction with therapy felt that they had improved significantly compared to the placebo group in their overall acrophobia symptoms ().
FIGURE 2 Reduction in acrophobia in the real world with DCS augmentation of virtual reality therapy. Assessment scores of acrophobia measures are shown at 3-month follow-up58
Our data indicate that participants receiving D-cycloserine experienced no increase in anxiety or fear during the exposure paradigm so that the enhancement of extinction is not due simply to enhanced intensity of exposure. Participants in the D-cycloserine group showed some evidence of enhanced extinction after only a single dose of medication and therapy. Following two doses of medication and therapy, they showed significant reductions in levels of fear to the specific exposure environment. Finally, we found that 3 months following the two treatment sessions, the D-cycloserine participants showed significant improvements on all acrophobia outcome measures, their own self-exposures in the real world, and their impression of clinical self-improvement.
Although it is possible that D-cycloserine somehow specifically enhances extinction, the current literature would suggest that it enhances learning in general, and thus enhances extinction as a form of learning. The specific evidence that D-cycloserine is enhancing extinction in a learning-specific way again comes from preclinical evidence in rodents. When combined with the conditioned stimulus, the D-cycloserine-treated animals showed accelerated extinction. However, this reduction was not seen when the animals were simply placed back in the fear conditioning context in the absence of the conditioned stimulus. Thus, D-cycloserine did not reduce fear by itself, but only facilitated the specific process of extinction of fear in combination with the exposure.59,62