Analysis of Neural Similarity Between Hypnotizer and Hypnotized Individuals Using Autoencoders and PCA

Hypnosis is a state of intense focus and heightened suggestibility, where individuals become absorbed in internal experiences like thoughts, emotions, and mental imagery. Research on neural oscillations has shown distinct brain activity patterns during hypnosis, including increased theta power in the left parietal and occipital regions, elevated beta power in the frontal and left temporal areas, and enhanced slow-gamma power in the frontal and left parietal regions. Most studies focus on the physiological and neural responses of the subjects under hypnosis, but the dynamic between the hypnotist and the subject, particularly how the hypnotist influences the subject’s brain activity, remains underexplored.In this study, EEG data was analyzed using a deep learning model (U-Net). After preprocessing, the cleaned EEG data were transformed into images with a resolution of 128×128×3. They were randomized and classified into train and test groups (0.8 train, 0.2 test) after normalization.Features extracted from the model’s bottleneck layer were examined with principal component analysis (PCA) to compare the EEG data of a hypnotist and a hypnotized individual. The results showed highly similar EEG representations at the bottleneck layer, confirmed by PCA, with only minor differences.These findings suggest that hypnosis may involve shared neural mechanisms, possibly supporting theories of neural synchronization or cognitive mirroring. While much is known about the brain activity of hypnotized individuals, research on the hypnotist’s neural dynamics is limited. Future research should involve simultaneous neuroimaging or EEG recordings of both the hypnotist and the hypnotized individual to explore whether hypnotists modulate brain activity intentionally and how these patterns interact, including functional connectivity measures like coherence or phase-locking.