Much of the biology underlying mental illness is poorly understood. Lack of knowledge about how the brain malfunctions is one reason so many mental disorders remain so frustratingly hard to treat. Thomas Insel, director of the National Institute of Mental Health, addressed this shortcoming—and the resulting plateau it has caused in new treatments—in his blog. The post, “A New Approach to Clinical Trials,” served as Insel’s introduction of a new NIMH policy.

According to these new rules, studies looking for funding would not only have to look at whether an intervention alleviates symptoms—depression or anxiety, for example—but also must serve as a scientific “probe” into how the disorder operates in the brain.

Clinical researchers would first define a biological target. This might be a receptor and its neurotransmitter, a certain neural circuit or a cognitive process. The study would then explore how that target is affected both by the mental illness itself and by the specific treatment under investigation. At the same time, researchers would study the treatment’s effects on outward symptoms. “Even if the treatment trial is negative—if there is no benefit in terms of treatment—we still will learn something about the disease,” says Nitin Gogtay, associate director for clinical research at NIMH.

Some researchers applaud this as a needed change. Jordan Smoller, director of the Psychiatric and Neurodevelopmental Genetics Unit at Massachusetts General Hospital, believes that NIMH’s biological approach could lead to a deeper understanding of how these disorders operate: “Those then become new opportunities for developing treatments.”

But not everyone is enthusiastic. The brain remains an uncharted labyrinth of neuronal connections. To demand that researchers specify a target for treatments such as cognitive behavioral therapy or exercise may be asking for a level of specificity that is just not possible, some researchers say. Even drug trials pose challenges. “There are very few molecules that somebody can ingest that hit a single target,” says Andrew Nierenberg, director of the Bipolar Clinic and Research Program and the Clinical Research Support Office at MGH. “Even if you do, you’re hitting a target that’s embedded in a complex biological network that then causes a whole cascade of other changes.”

Researchers also worry that this new focus eliminates a more natural bottom-up approach, where clinical observation leads to a trial, which then leads to a breakthrough treatment with a wider application. “People are suffering now. Keeping all avenues open, including serendipity and clinical observation and clinical wisdom, makes sense. This way you can have clinical trials arise from the field and not be superimposed from above,” Nierenberg says.

NIMH says it is taking these concerns into account. Gogtay admits that not every intervention will have a measurable brain target observable in an MRI or a PET scan. The new guidelines, he says, really define targets more broadly. And regarding researchers’ fears that bottom-up clinical observations will never be funded, Gogtay says, “There’s no reason why those trials will not be funded as long as they make a good case, at least make an effort of using them as scientific probes. It should be doable for any intervention.”

The policy is itself an effort to see if such guidelines can spur a better understanding of how mental illness affects the brain. Gogtay says NIMH will revisit the results in the next two to three years to see what insights about fundamental biological mechanisms the policy has spurred.