System and method for adaptive perceptual learning

• US 9,299,265 B2
• Filed: 11/30/2005
• Issued: 03/29/2016
• Est. Priority Date: 12/14/2001
• Status: Active Grant

• Current Assignee: Kellman Act Services, Inc.
• Original Assignee: Kellman Act Services, Inc.
• Inventors: Kellman, Philip J.
• Primary Examiner(s): Yao, Sam
• Assistant Examiner(s): CARLOS, ALVIN LEABRES
• Application Number:
  US11/291,286
• Publication Number:
  US 20060078856A1
• Time in Patent Office: 3,772 Days
• Field of Search: 434/118, 434/236, 434/350, 434/353, 434/322, 434/323, 600/300
• US Class Current: 1/1
• CPC Class Codes: G09B 5/00 Electrically-operated educational appliances working with questions and answers G09B7/00; simulators G09B9/00; advertising or displaying in general G09F
• CPC Class Codes: G09B 7/00 Electrically-operated teaching apparatus or devices working with questions and answers mechanically operated G09B3/00; computing arrangements G06F

Abstract

An adaptive learning system and method (“ALS”) for optimized, automated learning is disclosed. The optimal sequencing method is adaptive in the sense that it continuously monitors a student'"'"'s speed and accuracy of response in answering a series of questions, performing a series of classification tasks, or performing a series of procedures, and modifies the sequencing of the items presented as a function of these variables. One goal of the technique is to teach the subject matter in the shortest possible time. The optimal sequencing method may be used independently or in conjunction with disclosed perceptual learning and hinting methods.

Claims

1. A perceptual learning system for teaching students to recognize diagnostic structure from varying instances, the system comprising:

a microprocessor;

a memory accessible by the microprocessor;

an interface coupled to the microprocessor for interacting with a student;

a problem database in the memory including;

a plurality of learning items which are examples of different concepts including at least one given concept to be learned by the student, the different concepts being categories of structure that are distinguishable from one another by perceptual structure;

with respect to the concept to be learned, a plurality of learning items that are positive instances of the concept to be learned containing diagnostic structural elements sufficient to determine the concept to be learned that can be perceived by the student and a plurality of negative instances which do not contain diagnostic structural elements sufficient to determine the concept to be learned; and

with respect to the positive instances of the concept to be learned, a plurality of learning items corresponding to different examples of the concept to be learned that vary in attributes other than the diagnostic structural elements sufficient to determine the concept to be learned;

wherein the microprocessor is configured to present, in a presentation sequence and with a presentation delay interval, the learning items to the student as a series of learning trials, the series of learning trials including positive instances and negative instances of the concept from the problem database;

wherein the microprocessor is configured to vary the set of positive instances presented to the student in the series of learning trials, by retrieving and presenting to the student over the series of learning trials multiple different examples of positive instances of the concept to be learned that vary in irrelevant attributes other than the diagnostic structure sufficient to determine the concept,

and wherein the microprocessor is configured to receive a response from the student on at least some of the learning trials presented to the student; and

a delay counter for counting number of learning trials for each learning item presented to the student since last presentation of said each learning item;

wherein the microprocessor is configured to monitor the responses received from the student and determine a response speed and a response accuracy of the student'"'"'s response to one or more of the learning trials;

wherein the microprocessor is configured to vary the presentation sequence of the learning items presented to the student relating to the concept to be learned based upon the determined response speed and response accuracy, and

wherein the microprocessor is configured to vary the presentation delay interval for a learning item, based on the number of learning trials counted by the delay counter, the determined response speed, and the determined response accuracy.

2. The perceptual learning system of claim 1, wherein for the given concept to be learned, the problem database contains numerous instances not belonging to that concept, for use as negative instances in learning trials.

3. The perceptual learning system of claim 1, wherein for the given concept to be learned, the microprocessor is configured to vary the negative instances of the concept in their attributes and dimensions and share with positive instances of the concept various attributes and dimensions other than the diagnostic structure determining the concept.

4. The perceptual learning system of claim 1, wherein the microprocessor is configured to access the problem database and present the learning items as a structure discovery trial in which the microprocessor is further configured to solicit the student to recognize the diagnostic structure that determines the concept to be learned in a particular learning domain.

5. The perceptual learning system of claim 1, wherein the microprocessor is configured to access the problem database and present the learning items as a structure discovery trial in which the microprocessor is configured to solicit the student to distinguish positive and negative instances of the concept to be learned.

6. The perceptual learning system of claim 1, wherein the microprocessor is configured to access the problem database and present the learning items as a structure discovery trial in which the microprocessor is configured to solicit the student to classify learning items rapidly and accurately.

7. The perceptual learning system of claim 1, wherein the microprocessor is configured to access the problem database includes learning trials as a mapping trial in which the microprocessor is configured to solicit the student to associate a structure in one representational format with the structure as it appears in one or more different representational formats.

8. The perceptual learning system of claim 1, wherein the microprocessor is configured to access the problem database and present learning items as a structure mapping trial in which the microprocessor is configured to solicit an identification of the diagnostic structure that determines the concept to be learned in a particular learning domain.

9. The perceptual learning system of claim 1, wherein the microprocessor is configured to access the problem database and present learning items as a structure mapping trial in which the microprocessor is configured to solicit an identification of positive and negative instances of the concept to be learned.

10. The perceptual learning system of claim 1, wherein the microprocessor is configured to access the problem database and present learning items as a structure mapping trial in which the microprocessor is configured to solicit the student to classify learning items rapidly and accurately.

11. The perceptual learning system of claim 1, wherein

the microprocessor is configured to access the problem database and present learning trials as one or more structure discovery trials and one or more structure mapping trials,

wherein for the structure discovery trials he microprocessor is configured to:

solicit the student to identify the diagnostic structure that determines the concept to be learned in a particular learning domain,

solicit the student to distinguish positive and negative instances of the concept to be learned, and

solicit the student to classify learning items rapidly and accurately, and

wherein for the structure mapping trials the microprocessor is configured to solicit the student to associate a structure in one representational format with the structure as it appears in one or more different representational formats.

12. The perceptual learning system of claim 1, wherein for the learning trials, the microprocessor is configured to solicit a pattern classification response or a pattern comparison response.

13. The perceptual learning system of claim 1, wherein the microprocessor is configured to give feedback to the student regarding his performance after each learning trial.

14. The perceptual learning system of claim 1, further including a trial record database, wherein the microprocessor is configured to store response data in the trial record database regarding the student'"'"'s response to each learning item and the concept of which it is an instance, the response data including accuracy of the student'"'"'s answer to each learning item, and speed of the student'"'"'s response to each correctly answered learning item.

15. The perceptual learning system of claim 1, wherein the microprocessor is configured to select each learning item at random from the problem database.

16. The perceptual learning system of claim 1, wherein the microprocessor is configured to select each learning item for a given learning trial based on the student'"'"'s performance in one or more earlier learning trials.

17. The perceptual learning system of claim 1, wherein the microprocessor is configured to select each learning item from the problem database so as to reduce repetition of specific instances of the concept.

18. The perceptual learning system of claim 1, wherein the microprocessor is configured to present multiple concepts to the student for learning in a single learning session.

19. The perceptual learning system of claim 18, wherein the microprocessor is configured to select each concept at random from a category database.

20. The perceptual learning system of claim 18, wherein the microprocessor is configured to select each concept for a given learning trial based on the student'"'"'s performance in one or more earlier learning trials.

21. The perceptual learning system of claim 18, wherein the microprocessor is configured to select each concept for a given learning trial in a predetermined order.

22. The perceptual learning system of claim 1,

wherein the microprocessor is configured to retire individual concepts from the problem database upon meeting a predetermined learning criterion for instances of that concept,

wherein the learning criterion is a function of the student'"'"'s accuracy in answering novel instances of that concept, and response speed for each correct answer, over a predetermined number of presentations of instances of that concept.

23. The perceptual learning system of claim 22, wherein contributions to concept retirement of learner speed, accuracy, and number of presentations meeting speed and accuracy criteria, are parameter adjustable.

40. The perceptual learning system of claim 1, wherein the series of learning trials comprises one or more passive learning trials, wherein the microprocessor is configured to give information to the student and the student is not required to produce a response.

42. The perceptual learning system of claim 1, wherein the series of learning trials includes a series of classification trials including one or more active learning trials, wherein the microprocessor is configured to:

provide the one or more active learning trials; and

receive a response regarding the one or more active learning trials from the student.

45. The perceptual learning system of claim 1, wherein, for one of the learning trials, the microprocessor is configured to simultaneously present one or more of the positive instances and one or more of the negative instances of the concept.

46. The perceptual learning system of claim 45, wherein the microprocessor is configured to receive a response by the student identifying the one or more positive instances of the concept.

47. The perceptual learning system of claim 1, wherein, for one of the learning trials, the microprocessor is configured to:

present one of the positive instances of the concept or one of the negative instances of the concept; and

receive a response by the student identifying a presence or absence of a characteristic of the positive instances.

48. The perceptual learning system of claim 1, wherein, for one of the learning trials, the microprocessor is configured to receive a response by the student identifying a presence or absence of a characteristic of the positive instances in each of a set of the positive and negative instances displayed simultaneously.

49. The perceptual learning system of claim 1, wherein the microprocessor is further configured to present the same example of a positive instance of the concept to be learned only once within the series of learning trials.

50. The perceptual learning system of claim 1, wherein the problem database includes learning items directed to a plurality of concepts to be learned over the same series of learning trials, and where the microprocessor is further configured to present to the user learning items directed to each of the plurality of concepts to be learned.

24. A perceptual learning system for teaching students to recognize diagnostic structure from varying instances, the system comprising:

a computer system including a processor, a visual display device, at least one input device, and one or more memories, wherein the processor is coupled to the visual display device, the input device, and the one or more memories;

the one or more memories comprising;

a problem database containing a plurality of learning items and program instructions for execution by the processor, the program instructions including;

presenting one or more learning items as a structure discovery trial or a structure mapping trial, the structure discovery trial requiring a student to recognize the diagnostic structure that determines a concept in a particular learning domain, distinguish a plurality of positive instances and a plurality of negative instances of the concept, and process distinguishing information rapidly, and the structure mapping trial requiring the student to associate a structure in one representational format with the structure as it appears in one or more different representational formats;

presenting the learning items to the student, in a presentation sequence and with a presentation delay interval, as a series of classification trials, wherein the series of classification trials includes a pattern classification response or a pattern comparison response,

the series of classification trials including positive instances and negative instances of the concept with each of the positive instances containing at least one aspect of the diagnostic structure corresponding to the classification that can be perceived by the student, and

varying a set of the positive and negative instances in attributes other than the diagnostic structure determining the concept,

wherein the problem database contains numerous instances exemplifying a given concept to be learned,

each given concept is distinguishable by perceptual structure;

receiving a response from the student on at least some of the learning trials;

counting number of learning trials for each learning item presented to the student since last presentation of said each learning item;

monitor the responses received from the student and determining a response based on speed and a response accuracy of the student'"'"'s response to one or more of the learning trials; and

wherein the presentation sequence of learning items is varied based upon the determined response speed and response accuracy, and

wherein the microprocessor is configured to vary the presentation delay interval for a learning item, based on the number of learning trials counted by a delay counter, the determined response speed, and the determined response accuracy.

25. The perceptual learning system of claim 24, wherein the program instructions further include providing feedback regarding a performance of the student after each classification trial.

26. The perceptual learning system of claim 24, further including a trial record database, and wherein the program instructions further include transmitting response data regarding a response by the student to each classification trial to the trial record database for storage thereon, the response data including an accuracy of the response by the student to each classification trial, and a speed of the response by the student on each correctly answered classification trial.

27. The perceptual learning system of claim 24,

wherein the program instructions further include retiring concepts from the problem database upon meeting a predetermined learning criterion,

wherein the learning criterion is a function of an accuracy of the student in answering the learning items for that concept, and a response speed for each correct answer, over a predetermined number of repetitions of each concept.

28. The perceptual learning system of claim 27, wherein contributions to concept retirement of learner speed, accuracy, and number of presentations meeting speed and accuracy criteria, are parameter adjustable.

29. The perceptual learning system of claim 26, wherein the program instructions further include selecting at least one of the learning items for each classification trial at random from the problem database.

30. The perceptual learning system of claim 26, wherein the program instructions further include selecting each learning item for a given classification trial based on a performance by the student in one or more earlier classification trials.

31. The perceptual learning system of claim 26, wherein the program instructions further include selecting each learning item from the problem database so as to reduce repetition of specific instances of the concept.

32. The perceptual learning system of claim 26, wherein the program instructions further include presenting multiple concepts to the student for learning in a single learning session.

33. The perceptual learning system of claim 32, wherein the program instructions further include selecting each concept at random from a category database.

34. The perceptual learning system of claim 32, wherein the program instructions further include selecting each concept for a given classification trial based on a performance by the student in one or more earlier classification trials.

35. The perceptual learning system of claim 32, wherein the program instructions further include selecting each concept for a given classification trial in a predetermined order.

36. The perceptual learning system of claim 26,

wherein the program instructions further include retiring individual concepts from the problem database upon meeting a predetermined learning criterion for instances of that concept,

wherein the learning criterion is a function of an accuracy of the student in answering novel instances of that concept, and a response speed for each correct answer, over a predetermined number of presentations of instances of that concept.

37. The perceptual learning system of claim 24, wherein the computer system is part of a network and the one or more memories are located on a server remote from the student.

41. The perceptual learning system of claim 24, wherein the series of classification trials comprises one or more passive learning trials, wherein the program instructions further include providing information to the student and the student is not required to produce a response.

43. The perceptual learning system of claim 24, wherein the series of classification trials comprises one or more active learning trials, wherein the program instructions further include:

providing the one or more active learning trials; and

receiving a response regarding the one or more active learning trials from the student.

38. A perceptual learning method for teaching students to recognize diagnostic structure from varying instances, comprising the steps of:

providing a problem database containing a plurality of learning items which are examples of different concepts that are categories distinguishable by perceptible structure, which, with respect to any one concept, includes a number of positive instances containing diagnostic structural elements of the concept that can be perceived by a student and a number of negative instances which do not contain the diagnostic structural elements of the concept;

presenting the learning items to a student in a presentation sequence and with a presentation delay interval using a processor coupled to a memory containing the problem database to the student as a series of learning trials, the processor further coupled to one or more interfaces for interacting with the student;

wherein the series of learning trials includes a plurality of positive instances and a plurality of negative instances of the concept with each of the positive instances containing at least one of the diagnostic structural elements of the concept,

wherein the problem database contains a set of instances including the plurality of positive instances and the plurality negative instances of the concept to be learned,

wherein receiving responses to the learning items from the student via the one or more interfaces presented to the student; and

using the processor for;

counting number of learning trials for each learning item presented to the student since last presentation of said each learning item;

monitoring the responses received from the student and determining a response speed and a response accuracy of the student'"'"'s responses to at least one of the learning items;

varying the presentation sequence of the learning items based upon the determined response speed and response accuracy;

varying the presentation delay interval for a learning item, based on the number of learning trials counted by a delay counter, the determined response speed, and the determined response accuracy; and

presenting the varied presentation of the learning items to the student.

39. The perceptual learning system of claim 38, wherein the series of learning trials comprises one or more passive learning trials, wherein the method further comprises giving, by the processor, information to the student and the student is not required to produce a response.

44. The perceptual learning system of claim 38, wherein the series of learning trials includes a series of classification trials including one or more active learning trials, wherein the method further comprises:

providing, by the processor, the one or more active learning trials; and

receiving, by the processor, a response regarding the one or more active learning trials from the student.