Article

Meta-learning attentional strategies in multidimensional reinforcement learning

2026-08-08

Abstract excerpt

<p>Reward learning in complex, multidimensional environments requires balancing abstractgeneralisation with stimulus-specific precision. Previous work suggests that humans initiallyrely on feature-based abstraction to support fast generalisation, before shifting toward moreprecise, object-based learning. We asked whether humans also generalise these attentionalstrategies to novel contexts. In a large-scale study (...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 58141bdd-89e1-5767-9716-ab7722b6365a
DOI: 10.31234/osf.io/fs8c4_v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Meta-learning attentional strategies in multidimensional reinforcement learningDOI 10.31234/osf.io/fs8c4_v1Read

Compositional generalization in multi-armed banditsSemantic proximity Read

Multi-Task Reinforcement Learning in HumansSemantic proximity Read

Environment-sensitive generalization and exploration strategiesSemantic proximity Read

Temporal and state abstractions for efficient learning, transfer and composition in humansSemantic proximity Read

Learning generalizable representations through efficient codingSemantic proximity Read

Learning generalizable representations through efficient codingSemantic proximity Read

Learning generalizable representations through efficient codingSemantic proximity Read

Learning to Attend Through Value-Based Hypothesis TestingSemantic proximity Read

When models matter: Environmental demand guides the arbitration between model-based and model-free controlSemantic proximity Read

The Value of AbstractionSemantic proximity Read

Generalizing to generalize: when (and when not) to be compositional in task structure learningSemantic proximity Read

An inductive bias for slowly changing features in human reinforcement learningSemantic proximity Read