Liyu Xia

This study shows adolescent and adult mice use both reinforcement learning and short-term memory strategies to learn a conditional odor association task, with no developmental difference in learning rates but increased reliance on perseverative strategies in males.

This paper uses artificial neural networks to estimate parameters and identify cognitive models without likelihood computation, demonstrating effectiveness for models with inter-trial dependencies.

The paper investigates assumptions about parameter generalization and interpretability in reinforcement learning models across different tasks and participants.

This study investigates how probabilistic reinforcement learning changes with age during adolescence and adulthood, finding increased learning rates and decreased exploration with age, and a link between testosterone and learning rates in mid-adolescence.

This study investigates how humans use hierarchical temporal and state abstractions to transfer knowledge across tasks, showing transfer effects not explained by simpler models.