Simulation based training system for measurement of cognitive load to automatically customize simulation content
Inventors
Beaubien, Jeffrey • Feeney, John • DePriest, William N. • Pappada, Scott
Assignees
Publication Number
US-11532241-B1
Publication Date
2022-12-20
Expiration Date
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
In one example embodiment of the invention, a simulation based training system is provided having a sensor that unobtrusively collects objective data for individuals and teams experiencing training content to determine the cognitive states of individuals and teams; time-synchronizes the various data streams; automatically determines granular and objective measures for individual cognitive load (CL) of individuals and teams; and automatically determines a cognitive load balance (CLB) and a relative cognitive load (RCL) measure in real or near-real time. Data is unobtrusively gathered through physiological or other activity sensors such as electroencephalogram (EEG) and electrocardiogram (ECG) sensors. Some embodiments are further configured to also include sociometric data in the determining cognitive load. Sociometric data may be obtained through the use of sociometric badges. Some embodiments further automatically customize the simulation content by automatically selecting content based on the CL of the individuals and teams.
Core Innovation
In one example embodiment, a simulation based training system unobtrusively collects objective data from physiological and other activity sensors, time-synchronizes the various data streams, automatically determines granular and objective measures for individual cognitive load (CL) of individuals and teams, and automatically determines a cognitive load balance (CLB) and a relative cognitive load (RCL) measure in real or near-real time. Data is gathered through physiological or other activity sensors such as electroencephalogram (EEG) and electrocardiogram (ECG) sensors and may include sociometric data obtained through sociometric badges. Some embodiments further automatically customize the simulation content by automatically selecting content based on the CL of the individuals and teams.
The disclosure addresses technical problems in training simulators including how to translate real-time raw sensor data into objective measures of user cognitive state, how to aggregate individual-level measures to the team level using team cognitive load models, and how to select or modify the next scenario or simulation content from a large database to maximize training effectiveness. The technical solution defines a data format to translate raw sensor data in real-time, provides individual and team cognitive load models (including TDNN-based individual models and team models incorporating RCL and CLB), and uses real-time quantified cognitive load data to provide feedback and dynamically customize training to keep learners within an objective Zone of Proximal Development (ZPD).
Claims Coverage
The patent includes three independent claims. I identified seven main inventive features extracted from the independent claims.
Processor based simulator comprising a user interface, a cognitive load measuring subsystem, and a simulation content customization subsystem
A processor based simulator comprising: a user interface, a cognitive load measuring subsystem, and a simulation content customization subsystem; the user interface configured to present a first simulation content to a user;
Physiological sensor configured to capture raw input and communicate to the cognitive load measuring subsystem
A sensor comprising a physiological sensor configured to capture a raw input from the user and communicate the raw input to the cognitive load measuring subsystem;
Cognitive load measuring subsystem with processor, input translator, and predefined individual cognitive load model
The cognitive load measuring subsystem comprising: a processor, an input translator configured to translate the input to a compliant input value, and a predefined individual cognitive load model configured to determine a cognitive load measure from the compliant input value;
Simulation content customization subsystem determines second simulation content based on the cognitive load measure and communicates it to the user interface
The cognitive load measuring subsystem configured to communicate the user cognitive load measure to the simulation content customization subsystem; the simulation content customization subsystem configured to determine a second simulation content based on the cognitive load measure and communicate the second simulation content to the user interface; the processor based simulator configured to communicate the second simulation content to the user interface as a customized simulation content;
Zone of Proximal Development ranges used to determine second simulation content
The cognitive load measuring subsystem further comprises a predefined range of measures for a Zone of Proximal Development (ZPD) for the user; and the simulation content customization subsystem determines the second simulation content based on a comparison of the cognitive load measure to the ranges of measures for the ZPD for the user whereby a second cognitive load measure is more likely to be within range of measures for the ZPD.
Predefined individual cognitive load model comprises a time-delay neural network (TDNN)
Wherein the predefined individual cognitive load model comprises a time-delay neural network (TDNN).
Team-level models for team cognitive load, relative cognitive load, and cognitive load balance
A team cognitive load model configured to determine a team cognitive load value from the cognitive load value and a second cognitive load value; a relative cognitive load model configured to determine a relative cognitive load value from the team cognitive load value and one or more of the cognitive load value and the second cognitive load value; and the simulation content customization subsystem further configured to determine the second simulation content based on the relative cognitive load value.
The independent claims collectively cover a processor-based simulator architecture with physiological and optional sociometric sensors, an input translation layer, predefined individual cognitive load models (including TDNN), team-level CL analytics (TCL, RCL, CLB), and a simulation content customization subsystem that selects or modifies content based on cognitive load measures and ZPD ranges.
Stated Advantages
Unobtrusive collection of objective physiological and sociometric data to determine individual and team cognitive states in real or near-real time.
Automatic time-synchronization of multiple data streams and automatic calculation of individual CL, team CL, CLB, and RCL.
Real-time customization of simulation content based on cognitive load measures to keep learners within a Zone of Proximal Development (ZPD).
Provision of real-time alerts to instructors and the ability for instructors or the system to dynamically modify training scenarios on the fly.
Enhanced effectiveness of post-training After Action Reviews (AARs) by using objective, time-synchronized cognitive load measures.
Cost-effectiveness in maximizing training effectiveness given the substantial costs of designing, scheduling, and conducting team training exercises.
Documented Applications
Simulation based training (SBT) in medical fields including emergency medicine, surgery, cardiology, and pediatrics, using simulators for skill acquisition and maintenance of certification.
Team training scenarios where individual and team cognitive load measures are used to modify scenario difficulty, team size, or tasks to maintain the ZPD.
Automatic selection or modification of the next scenario or simulation content from a database or learning management system based on measured CL, RCL, and CLB.
After Action Reviews (AARs) that integrate time-synchronized physiological, observer ratings, video, and event logs to analyze team performance and cognitive states.
Health monitoring and individual and team workload monitoring outside of simulators and real-time individual and team training.
ACLAMATE (Automated Cognitive Load Assessment for Medical Staff Training and Evaluation) as an example embodiment to unobtrusively measure cognitive workload of each team member in real-time and provide alerts to instructors.
Pilot testing in a simulated operating room and clinically relevant patient care crisis scenarios including loss of airway, venous air embolism, and hemorrhage.
Interested in licensing this patent?