When researchers want to evaluate the efficacy of new anxiety treatments, the traditional approach is to study how rats or mice behave in uncomfortable or stressful situations. Rodents shun brightly lit, open spaces, where, in the wild, they would become easy prey.
So their natural tendency in a test apparatus is to find areas that are poorly illuminated or close to walls. The longer a medicated animal spends in areas in which it is unprotected, the more effective the drug is judged to be in treating anxiety.
But the drugs that have resulted from this approach are not actually very good at making people feel less anxious. Neither patients nor their therapists consider the available options – including benzodiazepines like Valium and selective serotonin reuptake inhibitors like Prozac or Zoloft – as adequate treatments for anxiety.
After decades of research, some of the big pharmaceutical companies are raising the white flag and cutting back on efforts to develop new anti-anxiety drugs.
But we cannot afford to give up on treatment for the so-called anxiety disorders, which encompass problems related to both fear and anxiety. Feelings of fear occur when a possible source of harm is nearby or likely to present itself, while feelings of anxiety usually involve the possibility of harm in the future.
Worldwide, the lifetime prevalence of anxiety disorders is about 15%, and the cost to society is enormous. In the late 1990s, it was estimated that the economic burden of anxiety totalled more than $40bn. The total cost is most likely significantly higher, because many anxiety disorders are never diagnosed.
Counter-intuitively, the reason that the most frequently prescribed anxiety medications don’t address the underlying problem is that they are working exactly as they should – according to the criteria used to design them.
Most treatments based on studies using mice or rats do make anxiety disorders easier to live with. What they fail to do is actually make people less fearful or anxious.
The reason for this is simple. The brain systems that control behavioural responses in threatening situations are similar in rodents and humans, and involve older areas deep in the brain that work nonconsciously (for example, the amygdala).
On the other hand, the systems that produce conscious experiences, including feelings of fear and anxiety, involve evolutionarily new regions of the neocortex that are especially well developed in the human brain and poorly developed in rodents.
Conscious feelings are also dependent on our unique linguistic capacities – our ability to conceptualise and name our inner experiences. It is telling that the English language has more than three dozen words for gradations of fear and anxiety: worry, concern, apprehension, unease, disquietude, inquietude, angst, misgiving, nervousness, tension and so forth.
Consequently, though animal studies are useful in predicting how a drug will affect nonconsciously controlled symptoms triggered by threatening stimuli, they are less effective when it comes to conscious feelings of fear or anxiety.
The drugs we have can help patients who, in order to avoid situations that inspire fear or anxiety, such as crowded subways or being judged by their peers or superiors, have stopped going to work.
Just as medicated rats are less behaviourally inhibited (more able to tolerate bright, open spaces), medicated anxiety sufferers are more likely to be able to return to their jobs. But, because the treatments do not directly address conscious brain processes, the anxiety itself does not always go away.
If treatments are to become more effective, our approaches will need to become more nuanced. We will need to treat the systems that operate nonconsciously differently from those that result in conscious experiences. This doesn’t necessarily mean better drugs.
Nonconscious responses can also be treated with exposure therapy, in which repeated interaction with a threatening stimulus is orchestrated in order to dampen its psychological effects.
Findings about how conscious and nonconscious brain systems work may enable us to make exposure therapy more effective. The basic idea is that the symptoms involving nonconscious processes should be targeted separately from those involving conscious processes.
I suggest the following sequence. Start with nonconscious exposures (using subliminal stimulation to bypass conscious thoughts and feelings that can be aroused and interfere with the exposure process) to dampen the response of areas like the amygdala.
Once the nonconscious systems are under control, use conscious exposures to treat conscious symptoms.
Finally, employ more traditional psychotherapies: Verbal interactions with the therapist aimed at helping patients work on changing beliefs, reevaluate memories, encourage acceptance of one’s circumstances, acquire coping strategies, and so on.
There is also a place for drugs in this approach, but not as a long-term solution. Rather, drugs can be used to make the exposure treatment more effective (the pharmaceutical d-cycloserine has shown some promise in this regard).
The effectiveness of an approach that recognises that different brain systems control different symptoms has yet to be properly evaluated, but research suggests that it should work.
It would also be noninvasive and would require only a repurposing of frequently used procedures. Given the magnitude of the problem, a stone so easily reached should not be left unturned. - Project Syndicate

- Joseph LeDoux, professor of science, neural science and psychology, and child and adolescent psychiatry at NYU, is director of the Emotional Brain Institute at the Nathan Kline Institute and NYU. His latest book is Anxious: Using the Brain to Understand and Treat Fear and Anxiety.