Predicting an EEG-Based hypnotic time estimation with non-linear kernels of support vector machine algorithm.

Area of Science:

• Cognitive Neuroscience
• Neuroscience of Time Perception
• Artificial Intelligence in Neuroscience

Background:

• Accurate time perception is crucial for daily functioning and mental well-being.
• Distinguishing time perception from other cognitive processes like emotion is a significant research challenge.
• Electroencephalography (EEG) offers a method to measure brain activity associated with cognitive functions.

Purpose of the Study:

• To investigate the effects of hypnosis on time estimation.
• To develop and evaluate artificial intelligence (AI) models for predicting time perception from EEG data.
• To identify brain activity patterns indicative of time underestimation and overestimation.

Main Methods:

• An experimental design combining hypnosis and EEG recordings.
• Utilizing hypnosis to reduce cognitive load and isolate time perception.
• Employing Support Vector Machines (SVMs) with radial basis function (RBF) and polynomial kernels for pattern classification.
• Analyzing various feature combinations and algorithms for optimal prediction accuracy.

Main Results:

• Hypnosis was used to modulate awareness and simplify cognitive processing during time estimation tasks.
• AI models, specifically SVM with an RBF kernel, demonstrated high accuracy in classifying time perception.
• An impressive 80.9% classification accuracy was achieved in predicting time perception from EEG data.

Conclusions:

• AI, particularly SVM with RBF kernel, shows significant potential in decoding brain activity related to time perception.
• This study highlights a novel approach combining hypnosis and AI for understanding the neural basis of time estimation.
• The findings open avenues for further research into the neurobiology of time perception and its clinical implications.