1 / 25

Distributed Cognitive Aid with Scheduling and Interactive Task Guidance

Distributed Cognitive Aid with Scheduling and Interactive Task Guidance. Edmund LoPresti, AT Sciences LLC Ned Kirsch, University of Michigan Debra Schreckenghost, Metrica TRACLabs Richard Simpson, University of Pittsburgh. Goals. Client.

bevan
Download Presentation

Distributed Cognitive Aid with Scheduling and Interactive Task Guidance

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Distributed Cognitive Aid with Scheduling and Interactive Task Guidance Edmund LoPresti, AT Sciences LLC Ned Kirsch, University of Michigan Debra Schreckenghost, Metrica TRACLabs Richard Simpson, University of Pittsburgh

  2. Goals Client Maintain autonomy of lifestyle in face of traumatic cognitive impairment ICue Caregiver Clinician - Determines techniques to counter impairment - Evaluates client condition and effectiveness of aids - Ensure client performs important tasks - Assist when client has problems

  3. Instruction Database Plan Database ICue Architecture Design Assistant Remote Information Server Activity Assistant Instruction Editor (Java) Cognition Manager HTN Planner for Schedules Reactive Planner for Instructions Plan Editor (Java) Clinician/Caregiver’s Office/Home Client’s Home

  4. Clinician Builds Task Instructions Design Assistant • Define steps of task • For each step, identify an instruction to aid client • Order the steps into a sequence • Define how client errors should be handled (alternative instructions, call in help) XML Instruction Database Clinician

  5. Caregiver Builds Client Plans Design Assistant • Identify activities for next day(s) • Add priorities and temporal constraints • ICue suggests a client plan • Adjust planning criteria to build a better plan, if desired Plan Database Caregiver

  6. Identify Activities for Schedule Task Attributes Available Tasks

  7. Add Priorities & Time Constraints • Timing • Required/preferred Priority Task List

  8. ICue Builds the Client Plan

  9. Review/Adjust Plan

  10. Task Stored as XML • Stored for later retrieval when a task becomes active • Transitioning from proprietary XML to AIMS-XML standard • Developed by AbleLink and RERC-ACT • Allow sharing between devices

  11. Instruction Database Plan Database Cognition Manager Activity Planner Instruction Sequencer ICue Tracks Schedule and Tasks Design Assistant XML

  12. Cognition Manager HTN Planner for Schedules Reactive Planner for Instructions ICue aids Client with Daily Tasks Activity Assistant • Remind client when time to perform task Step through task instructions • Provide alternative instructions when client has difficulty • Track whether client completes activities Client

  13. Preliminary Data: Field Trial • ICue prototype was used by a participant having cognitive impairment following TBI. • The participant was observed performing two individualized tasks with cues from ICue. • The participant was able to respond appropriately to the cues provided by ICue, and was able to successfully complete each task without intervention from a clinician, whereas the participant was unable to perform either task independently without the intervention.

  14. Preliminary Data:Usability Survey • 6 clinicians defined tasks for hypothetical clients • ICue seemed beneficial (mean score of 4.2 out of 5 for 2 questions) • Ease of use was mildly positive (mean scores of 3.7, 3.2, 3.7, and 3.7 for 4 questions); participant responses and investigator observations indicated usability issues to be addressed. • Some concern that ICue might take too long to use in a clinical setting (mean 2.5 out of 5). • Using ICue was not frustrating (mean score of 1.7 on a scale of 1-5 for frustration) and was worth the effort to use (mean 4.3 out of 5).

  15. Future Work: Feedback to Client/Caregiver • Real time (for situations ICue cannot handle automatically) • Logged data for later review; potentially increase or decrease level of cues based on client’s performance.

  16. Future Plans:Testing • Formal clinical testing • Usability trials with non-professional caregivers

  17. Future Plans:Internet Risk Factors • Loss of connectivity • Privacy

  18. Future Plans:Context Appropriate Cueing • Address behavioral issues • Provide simple prompts at regular intervals (e.g. pay attention, relax, speak more, speak less) • Provide support to vary these cues (wording and/or frequency) based on what is or is not appropriate in different contexts • A user’s schedule will be enhanced with information about the social context of their tasks • home, work, school; • alone, with friends, with co-workers; • etc

  19. Discussion / Question 1What are the biggest challenges/obstacles in automated task guidance? A. B. C. D. E.

  20. Discussion / Question 1What are the biggest challenges/obstacles in automated task guidance? A. B. C. D. E.

  21. Discussion / Question 2What technologies or techniques should be integrated to support caregivers? A. B. C. D. E.

  22. Discussion / Question 2What technologies or techniques should be integrated to support caregivers? A. B. C. D. E.

  23. Discussion / Question 3What broad categories of context might be important in behavioral cueing? A. B. C. D. E.

  24. Discussion / Question 3What broad categories of context might be important in behavioral cueing? A. B. C. D. E.

  25. Acknowledgements Work thus far: NIH SBIR #5 R43 HD44277-02 Environmentaly Appropriate Cueing RERC-ACT, NIDRR Grant #H133E040019

More Related