Computational Modeling of Cognitive Processes in Plan Authoring J. William Murdock & David W. Aha Intelligent Decision Aids Group Navy Center for Applied Research in Artificial Intelligence Naval Research Laboratory, Code 5515 Washington, DC 20375 {murdock,aha}@aic.nrl.navy.mil http://www.aic.nrl.navy.mil/~aha/ida/ TC3 Workshop: Cognitive Elements Of Effective Collaboration (January 15-17 2002, San Diego) Report Documentation Page Form Approved OMB No. 0704-0188 Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. 1. REPORT DATE 3. DATES COVERED JAN 2002 2. REPORT TYPE 00-00-2002 to 00-00-2002 4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER Computational Modeling of Cognitive Processes in Plan Authoring 5b. GRANT NUMBER 5c. PROGRAM ELEMENT NUMBER 6. AUTHOR(S) 5d. PROJECT NUMBER 5e. TASK NUMBER 5f. WORK UNIT NUMBER 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION Naval Research Laboratory,Navy Center for Applied Research in REPORT NUMBER Artificial Intelligence,Code 5515,Washington,DC,20375 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S) 11. SPONSOR/MONITOR’S REPORT NUMBER(S) 12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release; distribution unlimited 13. SUPPLEMENTARY NOTES TC3 Workshop: Cognitive Elements of Effective Collaboration, 15-17 Jan 2002, San Diego, CA. 14. ABSTRACT 15. SUBJECT TERMS 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF ABSTRACT OF PAGES RESPONSIBLE PERSON a. REPORT b. ABSTRACT c. THIS PAGE Same as 16 unclassified unclassified unclassified Report (SAR) Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18 Context Task: Plan authoring Task: Plan authoring • Intelligent system collaborates with human user • Intelligent system collaborates with human user • Human is in charge; system makes recommendations • Human is in charge; system makes recommendations Key Assumption: Key Assumption: • Human has a better “big picture” understanding of the • Human has a better “big picture” understanding of the situation than does the intelligent system situation than does the intelligent system Benefit: Benefit: • When the system makes a good recommendation, the • When the system makes a good recommendation, the user can accept rather than having to manually enter a user can accept rather than having to manually enter a new step in the plan. new step in the plan. • Faster plan authoring • Faster plan authoring Murdock & Aha, NRL 2/16 Challenge • An intelligent system can recommend actions for a plan • An intelligent system can recommend actions for a plan based on its limited knowledge. based on its limited knowledge. • We are considering situations in which the human has a • We are considering situations in which the human has a better general understanding of the problem. better general understanding of the problem. • The recommendation the system makes will, at first, be • The recommendation the system makes will, at first, be worse than what the human can do. worse than what the human can do. • How can we get the system to improve during the course • How can we get the system to improve during the course of the planning process? of the planning process? Murdock & Aha, NRL 3/16 Our Main Theme 1. Determine: 1. Determine: a. Goals: what the user is trying to do a. Goals: what the user is trying to do b. Approach: how the user is trying to do it b. Approach: how the user is trying to do it 2. Generate suggestions that are compatible with both 2. Generate suggestions that are compatible with both – i.e., Because the user is the expert, understand the user – i.e., Because the user is the expert, understand the user and then do things the user’s way. and then do things the user’s way. Goal: Travel from NRL to USD System requirement: A cognitive model of the user System requirement: A cognitive model of the user Murdock & Aha, NRL 4/16 Cognitive Models Task-Method-Knowledge (TMK): Task-Method-Knowledge (TMK): – Tasks: What a part of a process does. – Tasks: What a part of a process does. – Methods: How a part of a process works. – Methods: How a part of a process works. – Knowledge: What the process uses and alters. – Knowledge: What the process uses and alters. Existing work oMna TpMs, KR ohuatess ,involved: Existing work on TMK has involved: Travel – Intelligent sVyeshteicmless ,t heatct .automatically adapt – Intelligent systems that automatically adapt – Executable cognitive models of recorded protocols – Executable cognitive models of recorded protocols by Car by Train – Many other topics – Many other topics … – But never cognitive models of users – But never cognitive models of users Start Car Drive Stop Car Murdock & Aha, NRL 5/16 Overview of Decision Making Architecture TMK Analysis TMK Analysis Interpret the user’s reasoning and goals Interpret the user’s reasoning and goals TMK Prediction Automated Planning TMK Prediction Automated Planning Infer what the user intends to do next Select actions that accomplish the goals Infer what the user intends to do next Select actions that accomplish the goals Recommendation Filtering Recommendation Filtering Eliminate actions that are inconsistent with Eliminate actions that are inconsistent with the predictions of a user’s intentions the predictions of a user’s intentions Murdock & Aha, NRL 6/16 … Action Action Recommendation User Plan Authoring Tool Interpretation TMK of user’s TMK Expected Recommendation reasoning Behavior Analysis Prediction Filtering Inferred Goal Potential Actions TMK Cognitive Model of the User Automated Planner Knowledge Task Task Fact Operator Method Method Method … … Fact Operator Task Task Task Murdock & Aha, NRL 7/16 An Illustrative Logistics Example • User has some boxes to ship and two available trucks. • User has some boxes to ship and two available trucks. • One truck is faster, so a planner will recommend it. • One truck is faster, so a planner will recommend it. • The user wants to use the slower truck. • The user wants to use the slower truck. • Challenge: Recognize that the user has chosen the • Challenge: Recognize that the user has chosen the slower truck and make recommendations that abide by slower truck and make recommendations that abide by that choice. that choice. Murdock & Aha, NRL 8/16 What the user wants Fast Truck Boxes Boxes SSSlllooowww TTTrrruuuccckkk Slow Truck Slow Truck Slow Truck SSSlllooowww TTTrrruuuccckkk Murdock & Aha, NRL 9/16