Advances in understanding the basis of fear extinction offers new knowledge on anxiety disorders
Recent advances in understanding the behavioral, molecular, and anatomical bases of fear extinction in animals and humans are leading to new knowledge about the nature of fear and new treatments for anxiety disorders that affect millions of Americans.
Although every human being experiences some fear and anxiety during the course of normal life, excessive amounts of fear and anxiety are associated with almost all psychiatric illnesses. In those people who suffer from anxiety disorders like phobias, posttraumatic stress disorder, obsessive-compulsive disorder, and panic disorder, excessive and inappropriate fear and anxiety comprise the core symptoms of the disorder, says Mark Barad, PhD, of the University of California at Los Angeles and organizer of a symposium at this meeting on the extinction of conditioned fear.
The National Institute of Mental Health estimates that more than 19 million Americans suffer from such disorders. However, there are effective treatments for anxiety disorders, of which the most effective, termed behavioral therapies, are modeled explicitly on the extinction of conditioned fear in animals. Still, not all patients respond in the same way and better treatments are needed.
Conditioned fear describes the physical and emotional reactions elicited by sights, sounds, and smells associated with life-threatening or unpleasant experiences. As a child, everyone remembers their reluctance to get back on a bike after falling off and hurting themselves the first time. Fears waned, however, after you rode your bike again without falling off. This reduction of fear after repeated exposure to a fearful event without aversive consequences is known as extinction. Fear extinction can be likened to a form of inhibitory learning. You don’t erase your original fear, but instead you develop a new memory that competes with and suppresses the original fearful memory.
In earlier studies, Michael Davis, PhD, at Emory University found that a receptor for a particular protein called the N-methyl-D-aspartate (NMDA) receptor in a brain region called the amygdala is critical for the extinction of conditioned fear. In rat studies, a compound called D-cycloserine (DCS) into the amygdala was found to enhance the functioning of the NMDA receptor and to accelerate the rate of extinction.
Because findings about fear extinction show many similarities in rats and humans, Davis reasoned that combining DCS with psychotherapy could produce a more rapid and potentially more lasting reduction of phobias in humans. He and his colleagues found that acrophobics, patients who had an abnormal fear of heights, can overcome their fear of heights if they underwent psychotherapy that employed virtual reality.
Subjects were fitted with earphones and goggles that contained tiny television screens connected to a computer that simulated riding in a glass elevator in a large hotel. As the subjects moved their heads, the computer simulated views inside or outside the glass elevator. Their perspectives of the hotel changed as the elevator went up. Initially, most of these patients would only go up one or two floors. Beyond that point, it was simply too unpleasant for them. After about eight virtual reality sessions, however, most patients could remain on the elevator until they reached the highest floor.
To determine whether this fear extinction process could be accelerated, Davis and colleagues at Emory carried out a double-blind, placebo-controlled trial. Neither the therapist nor the patient knew whether they received single doses of DCS or sugar pills before each of two virtual reality sessions in the elevator. After one week and three months, they assessed the patients’ new level of anxiety in going up in the virtual elevator. Virtual reality therapy combined with DCS resulted in a more rapid and significant reduction of acrophobia compared to exposure therapy with placebo. This reduction of fear was maintained when the subjects encountered similar real-world situations. The people who had taken DCS reported an increased willingness to drive over high bridges or narrow mountain roads as well as go up in elevators. In fact, patients who received DCS with only two sessions of virtual reality therapy showed equal or greater improvement compared to patients who received eight sessions of the therapy.
Davis’ research supports the use of cognitive enhancers such as DCS, combined with psychotherapy, to accelerate the learning process of fear extinction. In addition, by reducing the needed duration of psychotherapy, DCS has the potential to lower the cost of treatment for various phobias as well as other psychiatric disorders.
In other studies, Gregory Quirk, PhD, of Ponce School of Medicine in Puerto Rico, has found an area of the brain that signals safety. In his experiments, Quirk found that rats easily learn to fear a tone that has been paired with an electric shock. If the tone is repeatedly presented without the shock, fear to the tone declines, a process called “extinction of fear.” Neurons in the medial prefrontal cortex of the brain are not activated during the initial fear learning, but are strongly activated after extinction, when fear is low.
“This appears to be an “all-clear” signal because rats that had the most activity in prefrontal cortex showed the least fear,” says Quirk. “Furthermore, mimicking this safety signal by stimulating prefrontal cortex eliminated fear in rats that were trained to expect a shock.” Quirk and his colleagues further found that prefrontal stimulation reduces fear because it reduces the activity of the amygdala, a locus of fear memory deep in the temporal lobe. Thus, fear can be turned off by mimicking the brain’s own safety signal.
For decades, it has been hypothesized that extinction does not erase the original fear memory, but forms a new memory of safety that inhibits fear expression. This suggests the presence of a brain region that increases its activity during extinction in order to signal safety and reduce fear. This structure now appears to be the medial prefrontal cortex, which is heavily interconnected with areas that generate fear, such as the amygdala and brainstem. Quirk’s findings suggest that prefrontal cortex stores a safety memory that allows us to stop fearing something that is no longer dangerous.
Most trauma victims do not develop PTSD, but are able to overcome their fears and lead a normal life. It has been suggested that PTSD sufferers are deficient in their ability to form strong extinction memories. The prefrontal region activated in rats during extinction is similar to an area in the human brain that has been shown to be atrophied in people with PTSD, supporting the idea the “all-clear” signal may be missing in these patients. These new findings could help develop future therapies aimed at strengthening this prefrontal safety-signal in people with PTSD and other anxiety disorders.
Elizabeth Phelps, PhD, of New York University, studies the mechanisms of extinction and active coping of fear in humans.
In one study, a basic fear conditioning paradigm was used to examine the neural mechanisms of extinction learning and the retention of extinction in normal subjects using fMRI. Subjects were presented with two colored squares, one blue and the other yellow. Subjects were told they would receive a mild shock to the wrist during some presentations of one of the squares, but not the other. Subjects received six shocks over fifteen presentations of each colored square. They demonstrated their learning of the association between the colored square and shock by showing greater arousal responses to the square paired with shock. After fifteen presentations of each colored square, the subjects did not receive any more shocks. They were given another 15 presentations of each square without any shocks on the same day (extinction day 1) , and 15 additional presentations the following day (extinction day 2).
Studies of extinction in nonhuman animals have highlighted two brain regions as important in extinction, the amygdala and prefrontal cortex. “Using brain imaging, we found evidence for involvement of both the amygdala and prefrontal cortex (primarily superior prefrontal cortex) in extinction of conditioned fear in humans. Consistent with animal models, the prefrontal cortex seemed to be particularly involved in the retention of extinction learning,” says Phelps.
A second study aimed to extend animal models of the extinction of fear learning by exploring the neural mechanisms of active coping. One mechanism by which humans can diminish a fear response is extinction, which can be characterized as passive learning through experience that the relationship between a conditioned and unconditioned stimulus has changed. However, humans can also take a much more active role in attempting to diminish a fear response. It is not uncommon in cognitive behavioral therapy for techniques of emotion regulation to be practiced in an effort to alter emotional reactions. One emotion regulation strategy is reappraisal, in which subjects actively shift their focus and appraisal of an emotional event. For instance, the common phrase “the glass is half full” reflects a reappraisal of a situation by focusing on the positive (rather than negative) aspects or consequences.
In the second study, subjects used a reappraisal strategy to diminish the fear response during a fear conditioning paradigm. The paradigm is similar to that described above, however, prior to conditioning subjects practiced generating images of soothing scenes from nature that were prompted by the colored squares. For instance, if a yellow square was presented, the subject might imagine a field of yellow daisies on a bright summer day. During the conditioning procedure, subjects were given an instruction to reappraise on some presentations of the colored squares, and on other trials were simply instructed to attend. During the reappraise trials, the subjects were told to focus on the soothing image prompted by the square, as opposed to the fact that the square might predict an aversive shock.
The reappraisal strategy was successful in that subjects showed less physiological arousal to the square that predicted shock on the reappraise (vs. attend) trails. In other words, the reappraisal strategy helped diminish the fear response. The fMRI results demonstrated both amygdala and prefrontal cortex involvement in reappraisal, much like extinction. However, the region of the prefrontal cortex most active in reappraisal was more lateral and the amygdala response was diminished on the reappraise trials
“The present results show both similarities and differences in the neural systems involved in diminishing fear responses through classic extinction and reappraisal,” says Phelps. “This research largely supports animal models of extinction. They also extend these models to include active coping as a mechanism for the extinction of fear learning.”
Mark Barad at UCLA has found several ways in which extinction of fear differs from other forms of learning, both in terms of learning rules and in terms of a molecular mechanism. He also is working on a drug that facilitates extinction, which may help improve the efficiency of psychotherapy, especially for human anxiety disorders.’
Barad and his colleagues study extinction using a conditioned fear condition by training mice to fear a noise on one day by pairing it with footshock, and then extinguishing the fear by presenting the noise many times the next day with no footshock. Fear is measured by watching the behavior of the mice. Mice that are not afraid are constantly on the move, walking and rearing and sniffing around in their cage, but when they are afraid, they “freeze,” becoming stiff and completely motionless, except for breathing. By counting the percentage of time a mouse spends freezing during a cue, the researchers can accurately estimate the fear of the mouse.
Because extinction is a form of learning, Barad began a set of studies with the hypothesis that extinction learning would resemble other forms of learning in its mechanisms, and began to test that hypothesis. Instead, he has found that extinction differs from other forms of learning in several interesting and possibly useful ways.
One of the most common learning rules is that temporally spaced training generates more and longer-lasting learning than massed training. Barad found that the opposite was true for extinction. When several cue presentations were massed only 5 seconds apart substantial extinction was generated, and that extinction lasted at least 8 days. However, when the same number of cue presentations were spaced 10 or 20 minutes apart no extinction occurred at all, either acutely or later.
“Our data indicate that cue presentations evoke two opposing processes, one that tends to strengthen fear in mice, and one that tends to extinguish it,” Barad says. “Somehow the spaced presentations favor the fear strengthening process, while massed presentations evoke the fear weakening learning of extinction. This finding has important implications for behavior psychotherapy of anxiety.
“It supports the idea that exposures to anxiety provoking situations, such as holding a spider for an arachnophobic or riding in elevators for someone afraid of elevators, will be most effective when they are prolonged; thus they appear to support the superior efficacy of “flooding” type exposures rather than incremental exposures.”
In the studying the molecular basis of extinction learning, Barad and his colleagues found an even more fundamental difference in the molecular basis of extinction from most forms of learning. While most forms of learning depend on the NMDA receptor both for their induction and for their consolidation, data from other researchers indicate that extinction depends on the NMDA receptor only during consolidation. Acute extinction proceeds normally in the presence of NMDA antagonists, but then does not persist to the next day. But what accounts for the initial extinction learning? By injecting inhibitors of L-type voltage gated channels into mice, Barad and his colleagues found that, in contrast to NMDA receptors, L-type voltage gated calcium channels are unnecessary for the acquisition of fear, but absolutely essential for all extinction learning. This provides substantial evidence that extinction differs in its fundamental molecular mechanisms from most forms of learning.
Finally, Barad and his colleagues have also have found that the adrenergic agonist, yohimbine, speeds up extinction with massed exposures, but most interestingly, it appears to overcome the fear-increasing influence of spaced cue presentations. Animals treated with yohimbine show significant extinction after spaced cue presentations that generate no extinction in controls. These data suggest that yohimbine may be a useful adjunct to psychotherapy.
Dawn McCoy | Society for Neuroscience