# Humans decompose tasks by trading off utility and computational cost

arXiv: https://arxiv.org/abs/2211.03890 · HTML: https://mlsanity.com/paper/2211.03890/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 53 references that are in the map.
- First posted on arXiv 7 Nov 2022.

## Builds on (most cited first, 7 of 7)

- [Planning with Goal-Conditioned Policies](https://mlsanity.com/paper/1911.08453/) (2019)
- [When Waiting is not an Option : Learning Options with a Deliberation Cost](https://mlsanity.com/paper/1709.04571/) (2017)
- [People construct simplified mental representations to plan](https://mlsanity.com/paper/2105.06948/) (2021)
- [Finding Options that Minimize Planning Time](https://mlsanity.com/paper/1810.07311/) (2018)
- [Towards a Quantum-Like Cognitive Architecture for Decision-Making](https://mlsanity.com/paper/1905.05176/) (2019)
- [Resource-rational Task Decomposition to Minimize Planning Costs](https://mlsanity.com/paper/2007.13862/) (2020)
- [KADABRA is an ADaptive Algorithm for Betweenness via Random Approximation](https://mlsanity.com/paper/1604.08553/) (2016)

## Built on by (most cited first, 18 of 18)

- [Explore, Select, Derive, and Recall: Augmenting LLM with Human-like Memory for Mobile Task Automation](https://mlsanity.com/paper/2312.03003/) (2023)
- [m&m's: A Benchmark to Evaluate Tool-Use for multi-step multi-modal Tasks](https://mlsanity.com/paper/2403.11085/) (2024)
- [Never Lost in the Middle: Mastering Long-Context Question Answering with Position-Agnostic Decompositional Training](https://mlsanity.com/paper/2311.09198/) (2023)
- [Learning Task Decomposition to Assist Humans in Competitive Programming](https://mlsanity.com/paper/2406.04604/) (2024)
- [Exploring the hierarchical structure of human plans via program generation](https://mlsanity.com/paper/2311.18644/) (2023)
- [AI for Mathematics: A Cognitive Science Perspective](https://mlsanity.com/paper/2310.13021/) (2023)
- [LiRA: A Multi-Agent Framework for Reliable and Readable Literature Review Generation](https://mlsanity.com/paper/2510.05138/) (2025)
- [CHOP: Mobile Operating Assistant with Constrained High-frequency Optimized Subtask Planning](https://mlsanity.com/paper/2503.03743/) (2025)
- [Evaluating Language Models' Evaluations of Games](https://mlsanity.com/paper/2510.10930/) (2025)
- [Learning Abstractions for Hierarchical Planning in Program-Synthesis Agents](https://mlsanity.com/paper/2602.00929/) (2026)
- [SG-FSM: A Self-Guiding Zero-Shot Prompting Paradigm for Multi-Hop Question Answering Based on Finite State Machine](https://mlsanity.com/paper/2410.17021/) (2024)
- [Emulating Human Cognitive Processes for Expert-Level Medical Question-Answering with Large Language Models](https://mlsanity.com/paper/2310.11266/) (2023)
- [Knowledge Model Prompting Increases LLM Performance on Planning Tasks](https://mlsanity.com/paper/2602.03900/) (2026)
- [Toward a normative theory of (self-)management by goal-setting](https://mlsanity.com/paper/2302.02633/) (2023)
- [Structurally guided task decomposition in spatial navigation tasks](https://mlsanity.com/paper/2310.02221/) (2023)
- [ChatGPT Unveils Its Limits: Principles of Law Deliver Checkmate](https://mlsanity.com/paper/2510.19261/) (2025)
- [Structure-Induced Information for Rerooting Levin Tree Search](https://mlsanity.com/paper/2605.30664/) (2026)
- [X-Tree: Tokenizing Reusable Experience for Efficient Agent Generalization](https://mlsanity.com/paper/2609.32993/) (2026)

## Abstract (arXiv)

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 provide new theoretical insight into the computational principles underlying the intelligent structuring of goal-directed behavior.