# Cognitive maps are generative programs

arXiv: https://arxiv.org/abs/2504.20628 · HTML: https://mlsanity.com/paper/2504.20628/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 45 references that are in the map.
- First posted on arXiv 29 Apr 2025.

## Builds on (most cited first, 13 of 13)

- [Translating Natural Language to Planning Goals with Large-Language Models](https://mlsanity.com/paper/2302.05128/) (2023)
- [WorldCoder, a Model-Based LLM Agent: Building World Models by Writing Code and Interacting with the Environment](https://mlsanity.com/paper/2402.12275/) (2024)
- [Combining Induction and Transduction for Abstract Reasoning](https://mlsanity.com/paper/2411.02272/) (2024)
- [People construct simplified mental representations to plan](https://mlsanity.com/paper/2105.06948/) (2021)
- [Centaur: a foundation model of human cognition](https://mlsanity.com/paper/2410.20268/) (2024)
- [Communicating Natural Programs to Humans and Machines](https://mlsanity.com/paper/2106.07824/) (2021)
- [People infer recursive visual concepts from just a few examples](https://mlsanity.com/paper/1904.08034/) (2019)
- [Learning abstract structure for drawing by efficient motor program induction](https://mlsanity.com/paper/2008.03519/) (2020)
- [Using Natural Language and Program Abstractions to Instill Human Inductive Biases in Machines](https://mlsanity.com/paper/2205.11558/) (2022)
- [Discovering State and Action Abstractions for Generalized Task and Motion Planning](https://mlsanity.com/paper/2109.11082/) (2021)
- [Exploring the hierarchical structure of human plans via program generation](https://mlsanity.com/paper/2311.18644/) (2023)
- [Map Induction: Compositional spatial submap learning for efficient exploration in novel environments](https://mlsanity.com/paper/2110.12301/) (2021)
- [Vector-based Pedestrian Navigation in Cities](https://mlsanity.com/paper/2103.07104/) (2021)

## Built on by (most cited first, 2 of 2)

- [Online library learning in human visual puzzle solving](https://mlsanity.com/paper/2603.23244/) (2026)
- [Prospective Compression in Human Abstraction Learning](https://mlsanity.com/paper/2605.09985/) (2026)

## Abstract (arXiv)

Making sense of the world and acting in it relies on building simplified mental representations that abstract away aspects of reality. This principle of cognitive mapping is universal to agents with limited resources. Living organisms, people, and algorithms all face the problem of forming functional representations of their world under various computing constraints. In this work, we explore the hypothesis that human resource-efficient planning may arise from representing the world as predictably structured. Building on the metaphor of concepts as programs, we propose that cognitive maps can take the form of generative programs that exploit predictability and redundancy, in contrast to directly encoding spatial layouts. We use a behavioral experiment to show that people who navigate in structured spaces rely on modular planning strategies that align with programmatic map representations. We describe a computational model that predicts human behavior in a variety of structured scenarios. This model infers a small distribution over possible programmatic cognitive maps conditioned on human prior knowledge of the world, and uses this distribution to generate resource-efficient plans. Our models leverages a Large Language Model as an embedding of human priors, implicitly learned through training on a vast corpus of human data. Our model demonstrates improved computational efficiency, requires drastically less memory, and outperforms unstructured planning algorithms with cognitive constraints at predicting human behavior, suggesting that human planning strategies rely on programmatic cognitive maps.