System and Method for Knowledge Verification Utilizing Biopotentials and Physiologic Metrics

Abstract

A system and method for knowledge verification utilizing biopotentials and physiologic metrics, which includes a computer-based device having stored thereon Probe, Relevant and Gallery image data, and a biopotential amplifier removably connected to a human subject via disposable Ag/Ag—Cl electrodes. Furthermore, the system comprises an analog-to-digital (A/D) converter to digitize said biopotential data for subsequent storage on said computer-based device, analysis software for discriminating said subject's event-related response to the exogenous stimuli, a visual display system comprising an LCD video monitor, and control software for presenting the Probe, Relevant and Gallery visual stimuli in a weighted, pseudo-random sequence which can be modulated by the outcome of said analysis software. Probe image data are not generally known to said human subjects but relevant to the knowledge to be verified; Relevant image data are generally known to said human subjects but not relevant to the knowledge to be verified; and Gallery image data are not generally known to said human subjects and not relevant to the knowledge to be verified. Said knowledge verification system can utilize parametric or non-parametric, e.g., artificial neural networks, analysis to provide an output of verification, or non-verification of knowledge of interest. Exemplary headband and electrode configurations optimized to produce the desired signals are disclosed.

1 . A knowledge verification method, comprising
 presenting, on a monitor operatively connected to a computer, probe, relevant, and gallery data to a subject in a statistically weighted, pseudo-random sequence through a control program operating on the computer;   recording on the computer, a time history of the presentation of the probe, relevant, and gallery data;   recording on the computer, while the subject is experiencing the presentation of the probe, relevant, and gallery data, electrical activity in the subject's brain that is measured by one or more electrodes operatively connected to the computer;   determining, through data analysis software operating on the computer, whether the electrical activity contains one or more p-300 responses; and   determining, through the data analysis software operating on the computer, whether any of the p-300 responses are correlated with the subject experiencing the probe data by comparing the occurrence of any p-300 responses to the time history of the probe, relevant, and gallery data presented to the subject.   
     
     
         2 . The method of  claim 1 , wherein the presenting probe, relevant, and gallery data to the subject takes place while the subject is in an isolation booth that substantially reduces exogenous distractions and extraneous electrical noise. 
     
     
         3 . The method of  claim 1 , wherein the presenting probe, relevant, and gallery data proceeds according to a knowledge tree where the broadest category of probe data is presented first and more specific categories of the same general category of probe data are only presented if broader categories evoke a p-300 response. 
     
     
         4 . The method of  claim 1 , wherein the probe, relevant, and gallery data make up one or more sets of stimulus files, and the presenting of probe, relevant, and gallery data further comprises:
 presenting the sets serially from broad divisions of information to information of progressively greater specificity to discover selected knowledge, and   presenting subsequent sets based on whether the stimulus files indicate recognition of a stimulus.   
     
     
         5 . The method of  claim 1 , wherein the data analysis software implements a neural network that employs at least three layers of neuron-like units. 
     
     
         6 . The method of  claim 1 , wherein the data analysis software implements a neural network that employs ensemble averaging. 
     
     
         7 . The method of  claim 1 , wherein the data analysis software implements a neural network that employs one or more weighted modulators between each layer of neuron-like units. 
     
     
         8 . A knowledge verification system, comprising
 one or more electrodes;   a computer operatively connected to the one or more electrodes and having instructions encoded thereon for:
 presenting probe, relevant, and gallery data to a subject in a statistically weighted, pseudo-random sequence; 
 recording a time history of the presentation of the probe, relevant, and gallery data; 
 recording, while the subject is experiencing the presentation of the probe, relevant, and gallery data, electrical activity in the subject's brain that is measured by the one or more electrodes; 
 determining whether the electrical activity contains one or more p-300 responses; and 
 determining whether any p-300 responses are correlated with the subject experiencing the probe data by comparing the occurrence of any p-300 responses to the time history of the probe, relevant, and gallery data presented to the subject. 
   
     
     
         9 . The system of  claim 8 , further comprising an isolation booth that substantially reduces exogenous distractions and extraneous electrical noise. 
     
     
         10 . The system of  claim 8 , wherein presenting probe, relevant, and gallery data proceeds according to a knowledge tree where the broadest category of probe data is presented first and more specific categories of the same general category of probe data are only presented if broader categories evoke a p-300 response. 
     
     
         11 . The system of  claim 8 , wherein the probe, relevant, and gallery data make up one or more sets of stimulus files, and the presenting of probe, relevant, and gallery data further comprises:
 presenting the sets serially from broad divisions of information to information of progressively greater specificity to discover selected knowledge, and   presenting subsequent sets based on whether the stimulus files indicate recognition of a stimulus.   
     
     
         12 . The system of  claim 8 , wherein determining whether the electrical activity contains one or more p-300 responses is achieved by employing a neural network having at least three layers of neuron-like units. 
     
     
         13 . The system of  claim 8 , wherein determining whether the electrical activity contains one or more p-300 responses is achieved by employing a neural network employing ensemble averaging. 
     
     
         14 . The system of  claim 8 , wherein determining whether the electrical activity contains one or more p-300 responses is achieved by employing a neural network having neuron-like units with one or more weighted modulators. 
     
     
         15 . A knowledge verification system, comprising
 one or more electrodes;   a computer operatively connected to the one or more electrodes and having instructions encoded thereon for:
 presenting probe, relevant, and gallery data to a subject in a statistically weighted, pseudo-random sequence; 
 recording a time history of the presentation of the probe, relevant, and gallery data; 
 recording, while the subject is experiencing the presentation of the probe, relevant, and gallery data, electrical activity in the subject's brain that is measured by the one or more electrodes; 
 determining whether the electrical activity contains one or more event-related potentials; and 
 determining whether any event-related potentials are correlated with the subject experiencing the probe data by comparing the occurrence of any event-related potentials to the time history of the probe, relevant, and gallery data presented to the subject. 
   
     
     
         16 . The system of  claim 15 , wherein presenting probe, relevant, and gallery data proceeds according to a knowledge tree where the broadest category of probe data is presented first and more specific categories of the same general category of probe data are only presented if broader categories evoke an event-related potential. 
     
     
         17 . The system of  claim 15 , wherein the probe, relevant, and gallery data make up one or more sets of stimulus files, and the presenting of probe, relevant, and gallery data further comprises:
 presenting the sets serially from broad divisions of information to information of progressively greater specificity to discover selected knowledge, and   presenting subsequent sets based on whether the stimulus files indicate recognition of a stimulus.   
     
     
         18 . A knowledge verification system, comprising
 one or more electrodes;   a computer operatively connected to the one or more electrodes and having instructions encoded thereon for:
 presenting probe, relevant, and gallery data to a subject in a statistically weighted, pseudo-random sequence; 
 recording a time history of the presentation of the probe, relevant, and gallery data; 
 recording, while the subject is experiencing the presentation of the probe, relevant, and gallery data, electrical activity in the subject's brain that is measured by the one or more electrodes; 
 determining whether the electrical activity contains one or more event-related potentials having a latency of approximately 250 ms to 900 ms; and 
 determining whether any of the event-related potentials having a latency of approximately 250 ms to 900 ms are correlated with the subject experiencing the probe data by comparing the occurrence of the event-related potentials having a latency of approximately 250 ms to 900 ms to the time history of the probe, relevant, and gallery data presented to the subject. 
   
     
     
         19 . The system of  claim 18 , wherein presenting probe, relevant, and gallery data proceeds according to a knowledge tree where the broadest category of probe data is presented first and more specific categories of the same general category of probe data are only presented if broader categories evoke an event-related potential having a latency of approximately 250 ms to 900 ms. 
     
     
         20 . The system of  claim 18 , wherein the probe, relevant, and gallery data make up one or more sets of stimulus files, and the presenting of probe, relevant, and gallery data further comprises:
 presenting the sets serially from broad divisions of information to information of progressively greater specificity to discover selected knowledge, and   presenting subsequent sets based on whether the stimulus files indicate recognition of a stimulus.