Apparatus and method for training using a human interaction simulator

• US 7,198,490 B1
• Filed: 11/23/1999
• Issued: 04/03/2007
• Est. Priority Date: 11/25/1998
• Status: Expired due to Fees

• Current Assignee: Johns Hopkins University
• Original Assignee: Johns Hopkins University
• Inventors: Olsen, Dale E.
• Primary Examiner(s): MOSSER, KATHLEEN MICHELE
• Application Number:
  US09/448,617
• Time in Patent Office: 2,688 Days
• Field of Search: 434/350, 434236-238, 434/167, 707/4, 345/352, 706/2, 706/11, 706/927
• US Class Current: 434/365
• CPC Class Codes: G09B 5/065 Combinations of audio and video presentations, e.g. videotapes, videodiscs, television systems
• CPC Class Codes: G09B 7/04 characterised by modifying the teaching programme in response to a wrong answer, e.g. repeating the question, supplying a further explanation
• CPC Class Codes: Y10S 345/956 Language driven animation

Abstract

A computer based training tool and method that emulates human behavior using a computer-simulated person in a realistic scenario. It provides an interactive experience in detecting deception during interviews and acceptance of statements during interpersonal conversations. The simulated person provides verbal responses in combination with an animated video display reflecting the body language of the simulated person in response to questions asked and during and after responses to the questions. The questions and responses are pre-programmed and interrelated groups of questions and responses are maintained in dynamic tables which are constantly adjusted as a function of questions asked and responses generated. The system provides a critique and numerical score for each training session.

Claims

1. An interactive apparatus, comprising:

a plurality of video vignettes simulating a person;

a plurality of statements to be selected by a user of said apparatus;

a plurality of audio responses for articulation by said simulated person; and

a logic means for interrelating each of said statements to be selected by the user, said audio responses and said video vignettes, such that each of said statements can have a plurality of different audio responses and video vignettes associated therewith, said logic means comprising;

an emotional component comprising an emotional model of said simulated person, said emotional model determining the direction and magnitude of change between a plurality of emotional states of said simulated person in response to said statements selected by the user thereby affecting the selection from a list of available audio responses and video vignettes of one of said plurality of audio responses and one of said plurality of video vignettes in response to user selected ones of said plurality of statements; and

a logical component for tracking said audio responses and said video vignettes and determining said list of available audio responses and video vignettes that is consistent in light of previously selected ones of said plurality of audio responses and ones of said plurality of video vignettes, said logical component further assigning a probability to each available response on said list based on said simulated person'"'"'s emotional state and, using said probabilities and a pseudo-random number generator, selecting one of said plurality of audio responses and one of said plurality of video vignettes from said list.

2. An interactive apparatus as defined by claim 1, further comprising means for establishing a performance score for the user of said apparatus as a function of a history of the selected ones of said plurality of statements.

3. An interactive apparatus as defined by claim 2, further comprising means for establishing a performance score for the user of said apparatus as a function of the sequence of selection of said plurality of statements.

4. An interactive apparatus according to claim 1, wherein said interactive apparatus is voice activated.

5. An interactive training method, comprising the steps of:

creating a plurality of video vignettes simulating a person;

creating a plurality of statements to be selected by a user of said apparatus;

creating a plurality of audio responses for articulation by said simulated person; and

creating logic means for interrelating each of said audio responses, said video vignettes and said statements to be selected by the user, such that each of said statements can have a plurality of different audio responses and video vignettes associated therewith, said logic means comprising;

a logical component for tracking said audio responses and said video vignettes and randomly selecting an audio response and video vignette associated therewith from a list of available audio responses and associated video vignettes determined by said logical component to be consistent in light of previously selected audio responses and video vignettes associated therewith; and

an emotional component comprising an emotional model of said simulated person, said emotional model determining the direction and magnitude of change between a plurality of emotional states of said simulated person in response to said statements selected by the user, said logical component assigning a probability to each available response on said list based on said simulated person'"'"'s emotional state and using said probabilities to affect the random selection from said list of one of said plurality of audio responses and one of said plurality of video vignettes in response to user selected ones of said plurality of statements, said random selection causing said simulated person to behave unpredictably thereby providing realistic interactive training for the user.

6. An interactive method according to claim 5, wherein the user articulates the statements selected.

7. An interactive apparatus, comprising:

a plurality of video vignettes simulating a person;

a listing of a plurality of statements to be verbalized by a user of said apparatus;

a plurality of audio responses for articulation by said simulated person; and

a logic means for interrelating each of said audio responses, said video vignettes and said statements to be verbalized by the user, such that each of said statements can have a plurality of different audio responses and video vignettes associated therewith, said logic means comprising;

a logical component for tracking said audio responses and said video vignettes and randomly selecting an audio response and video vignette associated therewith from a list of available audio responses and associated video vignettes determined by said logical component to be consistent in light of previously selected audio responses and video vignettes associated therewith; and

an emotional component comprising an emotional model of said simulated person, said emotional model determining the direction and magnitude of change between a plurality of emotional states of said simulated person in response to said statements selected by the user, said logical component assigning a probability to each available response on said list based on said simulated person'"'"'s emotional state and using said probabilities to affect the random selection from said list of one of said plurality of audio responses and one of said plurality of video vignettes in response to user verbalized ones of said plurality of statements.

8. An interactive apparatus as defined by claim 7, comprising means for establishing a performance score for the user of said apparatus as a function of the statements selected from said plurality of statements verbalized by the user.

9. An interactive apparatus as defined by claim 8, comprising means for establishing a performance score for the user of said apparatus as a function of the sequence of verbalization of said statements selected from said plurality of statements verbalized by the user.

10. An interactive apparatus, according to claim 7, wherein said interactive apparatus is voice activated by articulating the statements to be verbalized.

11. An interactive training method, comprising the steps of:

creating a plurality of video vignettes simulating a person;

creating a plurality of statements to be verbalized by a user;

creating means for recognizing verbalized ones of said plurality of statements;

creating a plurality of audio responses for articulation by said simulated person; and

creating logic means for interrelating each of said audio responses, said video vignettes and said plurality of statements to be verbalized by the user, such that each of said statements can have a plurality of different audio responses and video vignettes associated therewith, said logic means comprising;

a logical component for tracking said audio responses and said video vignettes and randomly selecting an audio response and video vignette associated therewith from a list of available audio responses and associated video vignettes determined by said logical component to be consistent in light of previously selected audio responses and video vignettes associated therewith; and

an emotional component comprising an emotional model of said simulated person, said emotional model determining the direction and magnitude of change between a plurality of emotional states of said simulated person in response to said statements selected by the user, said logical component assigning a probability to each available response on said list based on said simulated person'"'"'s emotional state and using said probabilities to affect the random selection from said list of one of said plurality of audio responses and one of said plurality of video vignettes in response to user verbalized ones of said plurality of statements, said random selection causing said simulated person to behave unpredictably thereby providing realistic interactive training for the user.

12. An interactive system, comprising:

memory means including a plurality of statements;

memory means for video presentation; and

keyboard means for selecting one of said plurality of statements in response to a visual cue from said monitor means or an audio cue, said video presentation capable of being one of a plurality of video presentations possible in response to said selected one of said plurality of statements, said selection of said one of said video presentations being controlled by;

a logical component for tracking said video presentations and randomly selecting a video presentation associated therewith from a list of available video presentations determined by said logical component to be consistent in light of previously selected video presentations associated therewith; and

an emotional component comprising an emotional model of a simulated person appearing in said video presentation, said emotional model determining the direction and magnitude of change between a plurality of emotional states of said simulated person in response to said statements selected by a user, said logical component assigning a probability to each available video presentation on said list based on said simulated person'"'"'s emotional state and using said probabilities to affect the random selection from said list of one of said plurality of video presentations in response to said selected one of said plurality of statements.

13. A computer readable medium that stores a program for providing interactive training for a user, said program comprising:

means for creating a simulated person in a plurality of vignettes;

means for creating a plurality of statements to be verbalized;

means for recognizing verbalized statements;

means for creating a plurality of audio responses for articulation;

means for creating logic means for interrelating said audio responses,

simulated person and statements to be verbalized, such that each of said statements can have a plurality of different audio responses and vignettes associated therewith, said logic means comprising;

a logical component for tracking said audio responses and said vignettes and randomly selecting an audio response and vignette associated therewith from a list of available audio responses and associated vignettes determined by said logical component to be consistent in light of previously selected audio responses and vignettes associated therewith; and

an emotional component comprising an emotional model of said simulated person, said emotional model determining the direction and magnitude of change between a plurality of emotional states of said simulated person in response to said statements selected by the user, said logical component assigning a probability to each available response on said list based on said simulated person'"'"'s emotional state and using said probabilities to affect the random selection from said list of one of said plurality of audio responses and one of said plurality of vignettes in response to user verbalized ones of said plurality of statements, said random selection causing said simulated person to behave unpredictably thereby providing realistic interactive training for the user.

14. A computer readable medium that configures a computer to perform an interactive training method, said method comprising the steps of:

simulating a person in a plurality of vignettes;

selecting statements to be selected by a user;

articulating audio responses by the simulated person; and

interrelating the statements selected by the user, the audio response and the simulated person, such that each of said statements can have a plurality of different audio responses and vignettes associated therewith, wherein said interrelating step comprises the step of using;

a logical component for tracking said audio responses and said vignettes and randomly selecting an audio response and vignette associated therewith from a list of available audio responses and associated vignettes determined by said logical component to be consistent in light of previously selected audio responses and vignettes associated therewith; and

an emotional component comprising an emotional model of said simulated person, said emotional model determining the direction and magnitude of change between a plurality of emotional states of said simulated person in response to said statements selected by the user, said logical component assigning a probability to each available response on said list based on said simulated person'"'"'s emotional state and using said probabilities to affect the random selection from said list of one of said audio responses and one of said plurality of vignettes in response to user selected ones of said statements, said random selection causing said simulated person to behave unpredictably thereby providing realistic interactive training for the user.

15. An interactive apparatus comprising:

a means for simulating a person in a plurality of vignettes;

a plurality of statements to be selected by a user of said apparatus;

a plurality of audio responses for articulation by said simulated person; and

logic means for interrelating each of said statements to be selected by the user, said audio responses and said simulated person, such that each of said statements can have a plurality of different audio responses and vignettes associated therewith, said logic means comprising;

a logical component for tracking said audio responses and said vignettes and randomly selecting an audio response and vignette associated therewith from a list of available audio responses and associated vignettes determined by said logical component to be consistent in light of previously selected audio responses and vignettes associated therewith; and

an emotional component comprising an emotional model of said simulated person, said emotional model determining the direction and magnitude of change between a plurality of emotional states of said simulated person in response to said statements selected by the user, said logical component assigning a probability to each available response on said list based on said simulated person'"'"'s emotional state and using said probabilities to affect the random selection from said list of one of said plurality of audio responses and one of said plurality of vignettes in response to user selected ones of said plurality of statements.

16. An interactive training method, comprising the steps of:

simulating a person in a plurality of vignettes;

creating a plurality of statements to be selected by a user of said apparatus;

creating a plurality of audio responses for articulation by said simulated person; and

creating logic means for interrelating each of said audio responses, said simulated person and said statements to be selected by the user, such that each of said statements can have a plurality of different audio responses and vignettes associated therewith, said logic means comprising;

a logical component for tracking said audio responses and said vignettes and randomly selecting an audio response and vignette associated therewith from a list of available audio responses and associated vignettes determined by said logical component to be consistent in light of previously selected audio responses and vignettes associated therewith; and

an emotional component comprising an emotional model of said simulated person, said emotional model determining the direction and magnitude of change between a plurality of emotional states of said simulated person in response to said statements selected by the user, said logical component assigning a probability to each available response on said list based on said simulated person'"'"'s emotional state and using said probabilities to affect the random selection from said list of one of said plurality of audio responses and one of said plurality of vignettes in response to user selected ones of said plurality of statements, said random selection causing said simulated person to behave unpredictably thereby providing realistic interactive training for the user.

17. An interactive apparatus comprising:

means for simulating a person in a plurality of vignettes;

a listing of a plurality of statements to be verbalized by a user of said apparatus;

a plurality of audio responses for articulation by said simulated person; and

a logic means for interrelating each of said audio responses, said simulated person and said statements to be verbalized by the user, such that each of said statements can have a plurality of different audio responses and vignettes associated therewith, said logic means comprising;

a logical component for tracking said audio responses and said vignettes and randomly selecting an audio response and vignette associated therewith from a list of available audio responses and associated vignettes determined by said logical component to be consistent in light of previously selected audio responses and vignettes associated therewith; and

an emotional component comprising an emotional model of said simulated person, said emotional model determining the direction and magnitude of change between a plurality of emotional states of said simulated person in response to said statements selected by the user, said logical component assigning a probability to each available response on said list based on said simulated person'"'"'s emotional state and using said probabilities to affect the random selection from said list of one of said plurality of audio responses and one of said plurality of vignettes in response to user verbalized ones of said plurality of statements.

18. An interactive training method comprising the steps of:

simulating a person in a plurality of vignettes;

creating a plurality of statements to be verbalized by a user;

creating means for recognizing verbalized ones of said plurality of statements;

creating a plurality of audio responses for articulation by said simulated person; and

creating logic means for interrelating each of said audio responses, said simulated person and said plurality of statements to be verbalized by the user, such that each of said statements can have a plurality of different audio responses and vignettes associated therewith, said logic means comprising;

a logical component for tracking said audio responses and said vignettes and randomly selecting an audio response and vignette associated therewith from a list of available audio responses and associated vignettes determined by said logical component to be consistent in light of previously selected audio responses and vignettes associated therewith; and

an emotional component comprising an emotional model of said simulated person, said emotional model determining the direction and magnitude of change between a plurality of emotional states of said simulated person in response to said statements selected by the user, said logical component assigning a probability to each available response on said list based on said simulated person'"'"'s emotional state and using said probabilities to affect the random selection from said list of one of said plurality of audio responses and one of said plurality of vignettes in response to user verbalized ones of said plurality of statements, said random selection causing said simulated person to behave unpredictably thereby providing realistic interactive training for the user.

19. An interactive apparatus, comprising:

means for simulating a person in a plurality of vignettes;

a plurality of statements to be selected by a user of said apparatus;

a plurality of audio responses for articulation by said simulated person;

and

a logic means for interrelating each of said statements to be selected by the user, said audio responses and said simulated person, such that each of said statements can have a plurality of different audio responses and vignettes associated therewith, said logic means comprising;

a logical component for tracking said statements, said audio responses and said vignettes and determining a list of available statements, audio responses and vignettes that is consistent in light of previously selected ones of said plurality of statements, ones of said plurality of audio responses and ones of said plurality of vignettes, said logical component further assigning a probability to each available audio response and vignette on said list based on an emotional state of said simulated person and, using said probabilities and a pseudo-random number generator, selecting one of said plurality of audio responses and one of said plurality of said vignettes from said list.