Sang‐Hun Lee

The paper demonstrates that relational structure constrains individual value estimates in visual working memory through experiments with biased feedback and probabilistic modeling.

The paper examines how the early visual cortex represents sensory and mnemonic orientations in separate subspaces while preserving geometric structure during working memory tasks.

The paper explains how attractor dynamics in working memory cause both stimulus-specific and decision-consistent biases in visual estimation.

The paper presents psychophysical experiments and a Bayesian model showing that past decisions with finer granularity have a stronger influence on future decisions.

This study compares two hypotheses about how humans use corrective feedback in perceptual decision-making, finding that treating feedback as evidence of the world's state better explains behavior than treating it as a reward.

The paper shows that boundary updating explains how previous stimulus history affects decision uncertainty and choice bias through behavioral, physiological, and neural evidence.

This study investigates how tDCS affects the human visual cortex by measuring changes in baseline activity and spatial tuning using fMRI.

The paper investigates the neural mechanisms behind repulsive bias in binary classification, finding that boundary-updating in specific brain regions, not sensory-adaptation, drives this bias.

The paper develops a Bayesian model for classification with bounded memory, showing it aligns with human behavior and neural activity in perceptual tasks.