In the Future, Brain Scans Might Be Able to Predict Who Will Benefit From Cognitive Behavioral Therapy

By
Young woman reclining on couch looking at psychologist
Photo: Laurence Mouton/© Corbis. All Rights Reserved.

Cognitive behavioral therapy, a form of therapy that is centered on helping people to recognize and break negative patterns of thinking, is more or less the gold standard when it comes to talk therapies. Meta-analyses have shown that there’s a very strong, robust base of evidence that CBT can help people who are suffering from anxiety, depression, and other psychological problems — maybe even fear of the dentist.

That doesn’t mean it’s perfect, though — despite all of this success, it still only works for about half of all those who try it, and researchers aren’t really sure why some patients respond while others languish without improvement. If we could tell in advance which patients will be unlikely to benefit from CBT, precious time and resources could be saved by offering them alternative forms of psychotherapy or drug treatment. 

We aren’t there yet, but an exciting new finding points in that direction. Researchers report in a new paper in Social Cognitive and Affective Neuroscience that they were able to predict from socially anxious patients’ brain activity whether they would respond positively to a course of CBT.

The researchers, led by Heide Klumpp at the University of Illinois at Chicago, recruited 32 men and women ages 18 to 55, all of whom had been previously diagnosed with social anxiety disorder. Some also had depression or panic disorder, and none were on meds or currently in therapy for their psychological problems.

Klumpp and her team started out by scanning the participants’ brains while they completed a task in which they were asked to look at strings of letters and press one specific keyboard key any time they saw the letter X and another whenever they saw the letter N — a test of their ability to concentrate. Crucially, sometimes an angry or scared face appeared in the background — socially anxious people are known to be very sensitive to and distracted by such socially threatening images — while other times, the faces had a less distracting neutral expression.

Within a week of the brain scanning, all of the participants began a 12-week course of once-a-week CBT. The key question was whether specific aspects of the participants’ brain activity in response to distracting images of social threat would be related to how well they responded to the course of CBT.

The exciting result is that brain activity during the task was related to the subsequent outcomes achieved through CBT. Specifically, the more the participants’ brain activity showed signs that they struggled to control the distracting effect of fearful or angry faces during the task (to get technical, they showed extra activity in the dorsal anterior cingulate cortex, a brain structure involved in managing conflicting attentional demands), the more they tended to benefit from the CBT — that is, the more their self-reported anxiety symptoms were decreased over the course of their CBT treatment. Levels of activity in other, mostly frontal, brain regions also correlated with success with CBT, such as the supplementary motor area — regions all involved in managing conflicting demands during times of distraction.

The implication is that patients who struggle more, at a neural level, with the distracting effects of social threat, such as a glimpse of a concerned frown or a disapproving glare, have the most to gain from CBT. Klumpp and her colleagues liken this finding to a “biomarker” for revealing who will be suitable for this form of psychotherapy. Their results, though preliminary (at a minimum, more research is needed with bigger samples and a control group), add to similar results for other diagnoses — for example, during the last few years related findings have emerged for depression and PTSD (posttraumatic stress disorder). In each case, brain-activity levels during a task featuring threatening stimuli have been related to later outcomes from CBT.

These are encouraging results that offer a tantalizing hint of what may one day soon become a routine approach to pre-therapy assessment. However, there are reasons not to get too carried away just yet. For one thing, this study looked at findings averaged across the group as a whole, so we’re not at the point yet where we can scan an individual’s brain and predict how well they’ll fare with CBT.

For another, while this study found that patients who struggle, at a neural level, to manage social threat benefited more from CBT, another recent study by the same group effectively drew the opposite conclusion — that patients whose brain activity suggested they were better able to manage social threat gained more from CBT. The difference may have to do with the fact that CBT is multifaceted: For example, it involves work on changing habits of thought, but often also includes intentional exposure to sources of fear. Either way, before this approach reaches the clinic, the researchers are going to have to untangle some important details.

A final note of caution: While scientists are apparently edging closer to using brain scanning as a way to predict who will benefit most from CBT, this is not to be confused with the mistaken idea that it’s possible to diagnose mental illness with a brain scan, as has been misleadingly touted by TV psychiatrist Daniel Amen and his chain of Amen Clinics, among others. On this point, mark the words of the American Psychiatric Association in its consensus statement published in 2012 (PDF): “There are currently no brain imaging biomarkers that are currently clinically useful for any diagnostic category in psychiatry.”

Dr. Christian Jarrett (@Psych_Writer), a Science of Us contributing writer, is editor of the British Psychological Society’s Research Digest blog. His latest book is Great Myths of the Brain.