Professor Woo Choong-wan's research team from the Department of Global Biomedical Engineering at Sungkyunkwan University, in collaboration with American researchers, has developed a model that directly reads the content and emotions of thoughts from brain activity patterns.
Professor Woo Choong-wan of the Department of Global Biomedical Engineering at Sungkyunkwan University (right) and Hong-ji Kim, a doctoral student. [Photo by Sungkyunkwan University]
Our brain is active every moment without rest. This brain activity manifests as 'thoughts' to us. Although the natural flow of thoughts may feel random, most often they contain emotions and are related to oneself or reflect internal desires and goals. Therefore, the content and emotional state embedded in the flow of thoughts can serve as important indicators of individual personality, cognitive characteristics, and mental health.
However, since the flow of thoughts occurs spontaneously without conscious constraints, simply asking people what they are thinking about at the moment can change the content of their thoughts, making research challenging.
The research team developed a model that predicts two main axes of thought?'self-relatedness' and 'positive-negative emotion'?using functional magnetic resonance imaging (fMRI) data and machine learning algorithms. They created personalized story stimuli containing various contents and emotions through one-on-one interviews with research participants, then analyzed brain activity patterns while participants read their own stories inside an MRI machine.
The developed predictive model not only successfully decoded participants' thoughts but also accurately predicted the natural flow of thoughts of an additional 199 people tested.
Professor Woo said, "Many researchers have attempted to decode thoughts from the brain, but very few groups have conducted research to read the intimate emotions contained within them. Our research team, which has studied human emotions for a long time, aims to obtain information helpful for mental health by reading the emotions embedded in the spontaneous flow of thoughts."
The results of this study were published in the prestigious journal Proceedings of the National Academy of Sciences (PNAS, IF 11.1).
© The Asia Business Daily(www.asiae.co.kr). All rights reserved.

