Interests

I’m interested in how humans think, learn, and make decisions, especially when we are short on computation and memory. I like to start from theory and build computational models (with symbolic reps, e.g. program induction) and test them through online experiments.

My PhD focuses on human resource rationality and intrinsic motivation, with a special interest in how these play out in real-world education.

Recent in my mind: not just how we learn and make decisions, but why we value certain outcomes in the first place. What shapes our sense of what’s worth pursuing? How are our utility functions, our goals or values, formed over time? How do we model long-term influences like evolution and culture (religion or shared belief systems), as well as more immediate ones like communication (social interaction and family story) and, of course, pedagogy?

Recent publications

Path-dependent program induction under resource constraints explains human sequence learning

Preprint (2026) — a preliminary version appeared at the Annual Conference of the Cognitive Science Society (CogSci 2025 Poster)
Preprint | Poster | Code

Effective explanations support planning under uncertainty

Annual Conference of the Cognitive Science Society (CogSci 2026 Oral)
Published version | Code | Talk

Agent-centric learning: from external reward maximization to internal knowledge curation

Reinforcement Learning Conference (RLC 2025 Oral)
Published version | Poster | Talk

Analogy making as amortised model construction

Reinforcement Learning Conference (RLC 2025 Oral)
Published version

The forest for the trees: global vs. local advice in human-AI interaction

Annual Conference of the Cognitive Science Society (CogSci 2025 Oral)
Published version | Code

From curiosity to competence: how world models interact with the dynamics of exploration

Annual Conference of the Cognitive Science Society (CogSci 2025 Poster)
Published version | Poster | Code

Harmonizing program induction with rate-distortion theory

Annual Conference of the Cognitive Science Society (CogSci 2024 Poster)
Published version | Poster

Predictive, scalable and interpretable knowledge tracing on structured domains

International Conference on Learning Representations (ICLR 2024 Spotlight)
Published version | Poster | Code | Press

The dynamic and structured nature of learning and memory

Conference on Cognitive Computational Neuroscience (CCN 2023 Poster)
Published version | Poster

CReBot: Exploring interactive question prompts for critical paper reading

International Journal of Human-Computer Studies (IJHCS 2022)
Published version