What part of the brain makes us unhappy and how to “retrain” it?

In modern realities, many suffer from obsessive anxious thoughts, about 31% of people. Most do not attach importance to the problem, but, unfortunately, everything is more serious than it seems. The symptom entails a lot of unpleasant consequences: from emotional burnout and neuroses to severe depression. How to find the cause and prevent complications? An international team of scientists led by Kim Jong-nu from the Center for Neuroscience Research in South Korea, together with colleagues from the University of Arizona and Dartmouth College in the United States, conducted a study to develop a predictive model of rumination using machine learning. results published in the journal Nature Communications.

Negative thinking, or rumination, is the tendency to constantly worry about the past or the future, associated with mistakes, doubts, or internal conflicts. It was important for researchers to understand what mechanisms in the brain are responsible for disturbing thoughts and how to prevent them.

It is already known that rumination is associated with a complex of brain regions called the ‘default mode network’ (DMN). This network is active when we are not engaged in a specific task and are immersed in our thoughts. However, not all DMN regions are equally affected on rumination Which of them plays a key role?

To answer this question, Kim Jong-nu’s team used functional magnetic resonance imaging (MRI), a method of measuring brain activity by measuring changes in blood flow in different areas of the brain. MRI is an advanced neuroimaging technology that allows examinations to be carried out safely, painlessly and non-intrusively. The researchers scanned healthy participants and patients with depression at rest, comparing the data with their self-reported levels of rumination. It was possible to collect detailed information about the work and activity of most DMN sites.

Then, based on the MRI data, a machine learning model was created to simulate brain activity and automatically predict the level of rumination. A model that mimics one region—the dorsal medial prefrontal cortex (dmPFC)—does the best job. This means that dmPFC is responsible for the formation and maintenance of negative thoughts. At the same time, it is important not only how intensively dmPFC works, but also how it communicates with other parts of the brain.

The program has been tested on people with a confirmed diagnosis. It turned out that she is also able to predict the level of depression according to MRI. This means that machine learning can help diagnose and treat depressive disorders.

The researchers hope that their model will allow them to better understand the mechanisms of rumination and develop new strategies to overcome it. They plan to expand the study to other mental disorders associated with negative thinking, such as post-traumatic stress disorder, obsessive-compulsive disorder and social anxiety.