Humans decompose tasks by trading off utility and computational cost
Authors:
Open Access
- 1 June 2023
- journal article
- research article
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 19 (6) , e1011087
Abstract
Human behavior emerges from planning over elaborate decompositions of tasks into goals, subgoals, and low-level actions. How are these decompositions created and used? Here, we propose and evaluate a normative framework for task decomposition based on the simple idea that people decompose tasks to reduce the overall cost of planning while maintaining task performance. Analyzing 11,117 distinct graph-structured planning tasks, we find that our framework justifies several existing heuristics for task decomposition and makes predictions that can be distinguished from two alternative normative accounts. We report a behavioral study of task decomposition (N = 806) that uses 30 randomly sampled graphs, a larger and more diverse set than that of any previous behavioral study on this topic. We find that human responses are more consistent with our framework for task decomposition than alternative normative accounts and are most consistent with a heuristic—betweenness centrality—that is justified by our approach. Taken together, our results suggest the computational cost of planning is a key principle guiding the intelligent structuring of goal-directed behavior.Keywords
All Related Versions
Funding Information
- John Templeton Foundation (61454)
- Air Force Office of Scientific Research (FA 9550-18-1-0077)
- Army Research Office (ARO W911NF-16-1-0474)
This publication has 36 references indexed in Scilit:
- Computational Rationality: Linking Mechanism and Behavior Through Bounded Utility MaximizationTopics in Cognitive Science, 2014
- Cognitive control over learning: Creating, clustering, and generalizing task-set structure.Psychological Review, 2013
- Bonsai Trees in Your Head: How the Pavlovian System Sculpts Goal-Directed Choices by Pruning Decision TreesPLOS Computational Biology, 2012
- A Neural Signature of Hierarchical Reinforcement LearningNeuron, 2011
- Hierarchically organized behavior and its neural foundations: A reinforcement learning perspectiveCognition, 2009
- Q-Cut—Dynamic Discovery of Sub-goals in Reinforcement Learning2002
- Normalized cuts and image segmentationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2000
- Between MDPs and semi-MDPs: A framework for temporal abstraction in reinforcement learningArtificial Intelligence, 1999
- Depth-first iterative-deepeningArtificial Intelligence, 1985
- Planning in a hierarchy of abstraction spacesArtificial Intelligence, 1974