Adaptive performance trainer
Inventors
Popovic, Djordje • Davis, Gene • Berka, Chris • Behneman, Adrienne • Raphael, Giby
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-9173582-B2
Publication Date
2015-11-03
Expiration Date
Abstract
Techniques for accelerating training through optimization of the psychophysiological state of the trainee are provided. These techniques include an adaptive performance training system configured to acquire, analyzed, display, and translate data that reflects the psychophysiological state of the user, including the electrical activity of the brain (EEG), the heart (EKG), the musculature (EMG), respiration and other parameters that characterize the state of the user in real-time. The system includes a plurality of feedback mechanisms for providing visual, auditory, and/or tactile feedback based on the current psychophysiological state of the user and for facilitating moving the user toward a goal psychophysiological state for performing a particular task and for optimizing performance of that task.
Core Innovation
The invention relates to accelerating learning of a task by real-time data acquisition and analysis of real-time psychophysiological signals obtained from a user. A monitoring module continuously monitors psychophysiological signal data collected from a plurality of sensors attached to the user during a pre-shot period of the task to be learned. The expertise profiler module analyzes the psychophysiological signal data to generate a current psychophysiological state of the user by computing one or more power spectral density indices from electroencephalographic activity and converting the power spectral density indices into probabilities that mental effort and engagement belongs to one of a plurality of classes using a discriminant function.
The system compares the current psychophysiological state to one or more goal states associated with expert performance of the task. Based on a difference between the current psychophysiological state and the one or more goal states, the system generates a set of control signals. A real-time feedback generator receives the control signals during the pre-shot period and generates at least one multimodality feedback signal to provide feedback to guide the user toward the one or more goal states.
The multimodality feedback comprises a tactile indication to the user of the user's current psychophysiological state at a pace that replicates the user's heartbeat. The tactile indication decreases as the user's current psychophysiological state approaches an ideal shooting state for the task to be learned. The invention is implemented as an adaptive portable performance training device worn by the user, delivering the multimodality feedback in real-time during the pre-shot period based on the control signals received from the expertise profiler module.
Claims Coverage
The document includes two independent claims, one system claim and one method claim. Both are centered on real-time psychophysiological monitoring during a pre-shot period, EEG power spectral density-based state estimation via discriminant-function class probabilities, comparison to expert goal states, and multimodality feedback generation including tactile feedback paced to replicate heartbeat and decreasing toward an ideal shooting state.
Real-time monitoring during a pre-shot period
Continuously monitor psychophysiological signal data from a plurality of sensors attached to a user during a pre-shot period of the task to be learned.
EEG power spectral density indices converted to class probabilities using a discriminant function
Analyze psychophysiological signal data to generate a current psychophysiological state by computing one or more power spectral density indices from electroencephalographic activity and converting the indices into probabilities that mental effort and engagement belongs to one of a plurality of classes using a discriminant function.
Comparing current state to expert goal states and generating control signals
Compare the current psychophysiological state to one or more goal states associated with expert performance of the task, and generate a set of control signals based on the difference between the current psychophysiological state and the goal states.
Multimodality feedback with tactile indication paced to replicate heartbeat and decreasing toward an ideal shooting state
Generate at least one multimodality feedback signal to provide feedback to guide the user toward the goal states, where feedback comprises a tactile indication of the user's current psychophysiological state at a pace replicating the user's heartbeat, and the tactile indication decreases as the user's current psychophysiological state approaches an ideal shooting state.
Across the independent claims, the coverage requires real-time continuous monitoring during a pre-shot period, EEG-based power spectral density indices converted into class probabilities via a discriminant function to form a current psychophysiological state, comparison to expert performance goal states, and multimodality feedback that includes a tactile indication paced to replicate the user's heartbeat and decreasing as the state approaches an ideal shooting state.
Stated Advantages
Documented Applications
Training contexts referenced include marksmanship training, golf training, and basketball free throw, including an interactive environment system for real-world field environments.
The tactile indication of the user's current psychophysiological state decreases as the user's current psychophysiological state approaches an ideal shooting state for the task to be learned.
Interested in licensing this patent?