Cognitive behavioral therapy (CBT), a widely used treatment for depression, helps individuals develop coping skills, reinforce healthy habits, and challenge negative thoughts. But can changing thought patterns and behaviors lead to lasting changes in the brain?

New research from Stanford Medicine suggests it can—when tailored to the right patients. In a study of adults with both depression and obesity, a challenging combination to treat, problem-solving therapy (a type of CBT) reduced depression in a third of the participants. These patients also experienced positive changes in their brain circuitry.

Remarkably, these neural changes appeared within just two months of therapy and could predict which patients would benefit from long-term treatment.

The findings add to growing evidence that matching depression treatments to the neurological profiles of patients—since these differ between individuals—increases the chances of success.

This concept is already routine in other medical fields. As Leanne Williams, PhD, professor of psychiatry and behavioral sciences and director of Stanford Medicine’s Center for Precision Mental Health and Wellness, explained, “If you had chest pain, your doctor would run tests—like an ECG or a heart scan—to find the cause and determine treatment. But in depression, we have no tests for what’s happening in the brain, so treatment selection is often trial-and-error.”

Williams and Jun Ma, MD, PhD, professor at the University of Illinois at Chicago, co-led the study, which was published in Science Translational Medicine on Sept. 4. Their work is part of a larger trial, RAINBOW (Research Aimed at Improving Both Mood and Weight).

Problem-solving therapy, the CBT approach used in the trial, is designed to enhance cognitive skills like planning, troubleshooting, and ignoring irrelevant information. Therapists help patients identify real-life problems, brainstorm solutions, and choose the best course of action. These skills rely on a network of neurons known as the cognitive control circuit.

Earlier research from Williams’ lab found that about 25% of people with depression experience dysfunction in their cognitive control circuits, either with too much or too little activity.

The new study focused on adults with both major depression and obesity—symptoms often linked to cognitive control issues. These patients tend to respond poorly to antidepressants, with a low success rate of 17%.

Of the 108 participants, 59 received a year of problem-solving therapy along with their usual care (medications, primary care visits), while the other 49 continued with their usual care alone. Brain scans (fMRI) were conducted at the study’s start and at two, six, 12, and 24 months. During the scans, participants performed tasks designed to engage their cognitive control circuits, allowing researchers to track changes in brain activity.

“We wanted to know if this specific therapy could modulate the cognitive control circuit,” said Xue Zhang, PhD, the study’s lead author.

Alongside the brain scans, participants completed surveys that measured problem-solving ability and depression symptoms.

As with any treatment, problem-solving therapy didn’t work for everyone, but 32% of participants saw their symptoms decrease by at least half—double the 17% response rate for antidepressants.

In the usual care group, cognitive control circuit activity declined over time, which correlated with worsening problem-solving ability. In contrast, the therapy group showed a reversal: decreased brain activity corresponded with better problem-solving. Researchers believe this reflects the brain learning to process information more efficiently.

“Before therapy, their brains were working harder. Now, they’re working smarter,” Zhang said.

Across both groups, overall depression improved, but therapy’s impact on “feeling everything is an effort”—a key symptom of cognitive impairment—stood out, highlighting its real-world benefits. Participants reported clearer thinking, resuming work, and re-engaging in social activities.

Importantly, brain scans showed changes in cognitive control circuit activity after just two months of therapy, signaling early brain plasticity. “Real-world problem solving is literally changing the brain in a couple of months,” Williams noted.

Moreover, these early changes predicted which patients would benefit long-term, offering a potential tool for precision psychiatry—using brain scans to match patients with the most effective treatments.

“This is advancing the science and transforming lives,” Zhang said.

Researchers from the University of Washington, University of Pittsburgh School of Medicine, and The Ohio State University contributed to the study.