SYSTEMS AMD METHODS FOR EVALUATING NEUROBEHAVIOURAL PERFORMANCE FROM REACTION TIME TESTS

• US 20100311023A1
• Filed: 05/07/2010
• Published: 12/09/2010
• Est. Priority Date: 05/07/2009
• Status: Active Grant

• Current Assignee: Pulsar Informatics, Inc., Trustees Of The University Of Pennsylvania (University of Pennsylvania)
• Original Assignee: Pulsar Informatics, Inc.
• Inventors: KAN, Kevin Gar Wah, MOTT, Christopher Grey, MOLLICONE, Daniel Joseph
• Granted Patent: US 8,794,976 B2
• Time in Patent Office: Days
• US Class Current: 434/258
• CPC Class Codes: G09B 19/0038 Sports

Abstract

Method are provided for evaluating reaction time data obtained from a stimulus-response testing system. One exemplary method comprises: obtaining reaction time data, the reaction time data comprising a plurality of reaction times, each reaction time comprising an estimate of a time required for a subject to respond to a corresponding stimulus event; assigning a weight to each reaction time in the reaction time data in accordance with a weighting function, the weighting function comprising a rule that defines a mapping between reaction times and corresponding weights; and determining a weighted reaction time metric based at least in part on a sum of the weights assigned to the reaction times in the reaction time data.

Claims

1. A method for evaluating reaction time data obtained from a stimulus-response testing system, the method comprising:

obtaining reaction time data, the reaction time data comprising a plurality of reaction times, each reaction time comprising an estimate of a time required for a subject to respond to a corresponding stimulus event;

assigning a weight to each reaction time in the reaction time data in accordance with a weighting function, the weighting function comprising a rule that defines a mapping between reaction times and corresponding weights; and

determining a weighted reaction time metric based at least in part on a sum of the weights assigned to the reaction times in the reaction time data.

2. A method according to claim 1 wherein at least a portion of the weighting function comprises a step that occurs at a threshold time, the threshold time defining a boundary between acceptable reaction times and undesirably long reaction times.

3. A method according to claim 2 wherein the threshold time is user-configurable.

4. A method according to claim 2 comprising setting the threshold time based at least in part on statistical characteristics of the reaction time data.

5. A method according to claim 2 comprising setting the threshold time based at least in part on statistical characteristics of historical reaction time data for the subject.

6. A method according to claim 1 wherein at least a portion of the weighting function comprises a step that occurs at a threshold time, the threshold time defining a boundary between acceptable reaction times and undesirably short reaction times considered to be false starts where the subject has initiated a response prior to actually perceiving a stimulus event.

7. A method according to claim 6 wherein the threshold time is user-configurable.

8. A method according to claim 6 comprising setting the threshold time based at least in part on statistical characteristics of the reaction time data.

9. A method according to claim 6 comprising setting the threshold time based at least in part on statistical characteristics of historical reaction time data for the subject.

10. A method according to claim 1 wherein the weighting function comprises:

a first step that occurs at a first threshold time, the first threshold time defining a first boundary between acceptable reaction times and undesirably short reaction times considered to be false starts where the subject has initiated a response prior to actually perceiving a stimulus event; and

a second step that occurs at a second threshold time greater than the first threshold time, the second threshold time defining a second boundary between acceptable reaction times and undesirably long reaction times.

11. A method according to claim 1 wherein at least a portion of the weighting function comprises a linear function having a slope which assigns relatively undesirable weights to relatively high reaction times and relatively desirable weights to relatively low reaction times.

12. A method according to claim 11 comprising setting the slope of the linear function based at least in part on statistical characteristics of the reaction time data.

13. A method according to claim 1 wherein at least a portion of the weighting function comprises an exponential function having an exponent term which assigns relatively undesirable weights to relatively high reaction times and relatively desirable weights to relatively low reaction times.

14. A method according to claim 1 wherein the weighting function comprises a plurality of portions:

a first portion comprising a first step that occurs at a first threshold time, the first threshold time defining a boundary between acceptable reaction times and undesirably short reaction times considered to be false starts where the subject has initiated a response prior to actually perceiving a stimulus event, the first portion assigning relatively undesirable weights to reaction times lower than the first threshold time and relatively desirable weights to reaction times greater than the first threshold time;

a second portion defined between a second threshold time greater than the first threshold time and a third threshold time greater than the second threshold time, the second portion assigning relatively undesirable weights to relatively high reaction times between the second and third threshold times and relatively desirable weights to relatively low reaction times between the second and third threshold times.

15. A method according to claim 14 wherein the weighting function comprises a third portion defined between the first and second threshold times, the third portion assigning, to reaction times between the first and second threshold times, weights that are at least as desirable or more desirable than the most desirable weights assigned by the first and second portions.

16. A method according to claim 15 wherein the weighting function comprises a fourth portion defined for reaction times greater than the third threshold time, the fourth portion assigning, to reaction times greater than the third threshold time, weights that are less desirable than the weights assigned by the second and third portions but are more desirable than the weights assigned by the first portion.

17. A method according to claim 14 comprising setting at least one of the first, second and third threshold times based at least in part on statistical characteristics of the reaction time data.

18. A method according to claim 14 comprising setting at least one of the first, second and third threshold times based at least in part on statistical characteristics of historical reaction time data for the subject.

19. A method according to claim 1 comprising evaluating the weighted reaction time metric by comparing the weighted reaction time metric to a threshold and determining whether the weighted reaction time metric is greater than, less than or equal to the threshold.

20. A method according to claim 19 wherein the threshold is user-configurable.

21. A method according to claim 1 comprising evaluating the weighted reaction time metric by obtaining one or more second weighted reaction time metrics determined on a basis of one or more corresponding second sets of reaction time data and comparing the weighted reaction time metric to the second weighted reaction time metrics.

22. A method according to claim 21 wherein the one or more second weighted reaction time metrics comprise a plurality of second weighted reaction time metrics and wherein comparing the weighted reaction time metric to the second weighted reaction time metrics comprises ranking the weighted reaction time metric in relation to the plurality of second weighted reaction time metrics.

23. A method according to claim 21 wherein the one or more second weighted reaction time metrics are determined based on one or more corresponding second sets of reaction time data associated with the subject.

24. A method according to claim 1 comprising evaluating the weighted reaction time metric by:

generating a cumulative probability distribution based on a plurality of second weighted reaction time metrics, each of the second weighted reaction time metrics determined based on a corresponding second set of reaction time data; and

assessing a percentile of the weighted reaction time metric relative to the cumulative probability distribution.

25. A method according to claim 24 wherein each of the second weighted reaction time metrics is determined based on a corresponding second set of reaction time data associated with the subject.

26. A method according to claim 1 comprising evaluating the weighted reaction time metric by:

defining a 1-dimensional feature map having one or more regions which span the range of possible values of the weighted reaction time metric, the regions defined by boundaries;

locating the weighted reaction time metric within the 1-dimensional feature map; and

classifying the weighted reaction time metric based on its location within the 1-dimensional feature map.

27. A method according to claim 1 comprising:

repeating assigning weights to each reaction time and determining a weighted reaction time metric a plurality of n times, each time with a different weighting function to obtain a corresponding plurality of n weighted reaction time metrics; and

evaluating the plurality of n weighted reaction time metrics by;

defining an n-dimensional feature map having one or more regions which span the n-dimensional range of possible values of the plurality of n weighted reaction time metrics, the regions defined by boundaries;

locating the n weighted reaction time metrics within the n-dimensional feature map; and

classifying the plurality of n weighted reaction time metrics based on their locations within the n-dimensional feature map.

28. A method according to claim 1 wherein the weighted reaction time metric is correlated with a neurobehavioural characteristic of the subject and wherein the method comprises evaluating the neurobehavioural characteristic by comparing the weighted reaction time metric to a threshold.

29. A method according to claim 1 wherein the weighted reaction time metric is correlated with an ability of the subject to perform an operational task and wherein the method comprises evaluating the ability of the subject to perform the operational task by comparing the weighted reaction time metric to a threshold.

30. A method according to claim 1 wherein the weighted reaction time metric is correlated with at least one of:

a behavioural characteristic of the subject and a medical condition of the subject and wherein the method comprises evaluating the behavioural characteristic or the medical condition of the subject by comparing the weighted reaction time metric to a threshold.

31. A method according to claim 1 wherein the reaction time data comprises one or more false start times, each false start time comprising a time at which the subject responded in an absence of a corresponding stimulus event.

32. A method according to claim 31 comprising identifying false start times within the reaction time data and assigning, to each identified false start time, a common reaction time value.

33. A method according to claim 1 wherein determining the weighted reaction time metric comprises normalizing the sum by dividing the sum by a total number of valid reaction times within the reaction time data.

34. A method according to claim 33 wherein, prior to determining the weighted reaction time metric, excluding reaction times within a particular reaction time exclusion range.

35. A method according to claim 34 wherein the particular reaction time exclusion range comprises a range between zero and a false start threshold time.

36. A system for evaluating reaction time data obtained from a stimulus-response testing system, the system comprising:

a controller connected to receive reaction time data, the reaction time data comprising a plurality of reaction times, each reaction time comprising an estimate of a time required for a subject to respond to a stimulus event, the controller configured to;

assign a weight to each reaction time in the reaction time data in accordance with a weighting function, the weighting function comprising a rule that defines a mapping between reaction times and corresponding weights; and

determine a weighted reaction time metric based at least in part on a sum of the weights assigned to the reaction times in the reaction time data.

37. A method for evaluating the ability of a subject to perform an operational task, the method comprising:

providing a stimulus-response test system comprising a stimulus output device for outputting stimulus events and a response input device for receiving responses from the subject;

obtaining reaction time data from the stimulus response test system, the reaction time data comprising a plurality of reaction times, each reaction time comprising an estimate of a time required for a subject to respond to a corresponding stimulus event;

assigning a weight to each reaction time in the reaction time data in accordance with a weighting function, the weighting function comprising a rule that defines a mapping between reaction times and corresponding weights;

determining a weighted reaction time metric based at least in part on a sum of the weights assigned to the reaction times in the reaction time data, the weighted reaction time metric correlated with the ability of the subject to perform the operational task.

38. A method according to claim 37 comprising evaluating the ability of the subject to perform the operational task by comparing the weighted reaction time metric to a threshold.