Prospective Compression in Human Abstraction Learning

TL;DR

This paper investigates how humans learn reusable abstractions in non-stationary environments, proposing that they anticipate future tasks and optimize their learning accordingly, unlike existing retrospective algorithms.

Contribution

It introduces the hypothesis that human library learning in non-stationary domains is prospective, and provides experimental evidence supporting this behavior over retrospective models.

Findings

Humans adapt their abstraction strategies based on non-stationary task structures.

Existing algorithms fail to replicate human prospective compression behavior.

Experimental data aligns with models of prospective, anticipatory abstraction 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.