Human-computer interface systems and methods

A method of predicting and using emotions of a user in a virtual reality environment, the method comprising:

• applying a plurality of physiological sensors to a user, wherein the plurality of physiological sensors includes one or more brain sensors on a user's scalp;
• receiving physiological sensor signals from the physiological sensors;
• preparing the physiological sensor signals for further processing by removing at least some of the noise and artifacts;
• processing the data to address fault tolerance and compensating for imprecisions in sensor positioning on the user's scalp;
• producing an emotion-predictive signal by utilizing an emotion database, wherein the emotion database has been developed based on empirical data from physiological sensors with known emotional states using deep learning;
• delivering the emotion-predictive signal to a virtual-reality system;
• and
• wherein the plurality of physiological sensors comprise a plurality of EEG sensors, and wherein the step of processing the data to address fault tolerance and compensating for imprecisions in sensor positioning on the user's scalp comprises:
• projecting positions of the EEG sensors in spherical coordinates onto a cartesian system to eliminate dependency on anatomical head size,
• assigning a two-dimensional spatial probability to the EEG sensors on the user's scalp based on each sensor's signal intensity,
• embedding an EEG signal at each time interval into a two-dimensional spatial probability for each EEG sensor to form frames and a frame-by-frame map of brain electrical activities,
• stitching the frames together in time to form a three-dimensional block, wherein time is the third dimension, and which represents a probability distribution of fluctuations in brain signals throughout time,
• aligning the three dimensional block to the correct anatomical position of the head, and
• outputting the three-dimensional block to the emotion database for emotion prediction.