Scientists Use Brain Scans to Decode Emotions in Personal Thoughts
-
Researchers developed methods to predict emotions in spontaneous thoughts using fMRI scans and machine learning algorithms.
-
Participants crafted personalized narratives reflecting their experiences and emotions. Their brain activity was recorded while reading these stories.
-
Machine learning models were trained on the fMRI data to decode the emotional dimensions of thoughts in real time.
-
Key brain regions involved included the anterior insula, midcingulate cortex, temporoparietal junction, and dorsomedial prefrontal cortex.
-
The models showed promise in decoding self-relevance and emotional valence not only during story reading but also during free thinking and rest.