Simulation of human decision making

1. A method for computer emulation of human decision making comprising:

defining a plurality of concepts related to a domain;

representing each concept in the computer as a concept oscillator output;

defining a plurality of situations related to the domain, each situation formed of at least one concept;

representing each situation in the computer as a situation oscillator output, wherein each situation and concept oscillator output is distinguishable from all other oscillator outputs;

inputting information to the computer representative of detected concepts;

comparing in the computer the detected concepts with the stored situations to determine if a situation has occurred.

2. The method of claim 1 wherein each concept oscillator operates in a first frequency range when the level of activation of its concept exceeds its concept threshold, the level of activation being indicated by at least one of frequency and amplitude of the concept oscillator output.

3. The method of claim 2 wherein each situation oscillator operates in a second frequency range when the level of activation of the situation oscillator exceeds its threshold, the degree to which each situation occurs being indicated by the level of activation of at least one of frequency and amplitude of the situation oscillator output.

4. The method of claim 3 further comprising the computer determining the situation that has the greatest level of activation above its threshold, wherein that situation is considered to be the detected situation.

5. The method of claim 4 wherein the first frequency range is the bandwidth of the alpha brain wave in humans, and the second frequency range is the bandwidth of the theta brain wave in humans.

6. The method of claim 5 further comprising defining associations between pairs of concepts, each activated concept causing a change in the level of activation of associated concepts; and adjusting at least one of the frequency and amplitude of each actuated concept oscillator based on its association with another concept.

7. The method of claim 6 further comprising spreading activation to a second concept oscillator because of the activation of a first concept oscillator to which the second oscillator is associated, wherein the level of activation of the second oscillator increases because of the activation of the associated first oscillator.

8. The method of claim 7 further comprising stopping the effect of said first concept oscillator on said second concept oscillator when the activation of the first concept oscillator stops.

9. The method of claim 6 further comprising increasing the oscillation frequency of a node oscillator as the level of activation of the node increases.

10. The method of claim 6 further comprising top-down activation of associated concepts wherein recognition of a situation causes an increase in the level of activation of associated concepts.

11. The method of claim 6 further comprising representing in the computer a system oscillator operating within the alpha bandwidth, wherein the concept oscillators which are not actuated oscillate in synchronization with the system oscillator at a low amplitude.

12. The method of claim 11 further comprising updating the level of activation of each concept and situation oscillator with each cycle of the oscillator.

13. The method of claim 12 further comprising providing an indication of unexpected occurrence when a concept is actuated which is inconsistent with an actuated situation.

14. The method of claim 12 further comprising taking an action in response to a situation.

15. The method of claim 1 wherein the relationship between concepts and situations are defined by a knowledge elicitation process comprising:

determining a list of situations and concepts related to those situations;

defining the situations as combinations of cues; and

determining a quantitative relationship between the cues and situations; and

creating an associative network of cues based on semantic relationships between concept nodes.

16. The method of claim 1 wherein a plurality of subject matter experts each define cues, situations, and associations relative to a domain, and further comprising:

forming a separate model from each expert's cues, situations, and associations;

if at least one of these models fails to recognize a situation for the inputted information:

determining the nodes representing concepts recognized by all experts;

summing the sums of the levels of activation for each of said nodes in each of the models; and

declaring a situation has occurred if said sum exceeds a predetermined threshold.