Hansem Sohn

This study shows that rhesus monkeys learn maze-solving tasks significantly faster in a physical environment than in a virtual one, with limited performance transfer between the two.

This paper introduces dSPMs to identify brain algorithms for scene prediction in macaques, finding a 'manifold-based simulation' algorithm in the DMFC.

The paper shows that prior expectations shape neural activity by increasing firing thresholds in neurons tuned away from expected directions, improving behavioral reliability.

The paper identifies simplex-shaped neural manifolds for number representation in the human medial temporal lobe and finds similarities with large language models.

The paper tests the hypothesis that mental simulations underlie physical inferences by analyzing neural activity in monkeys during a ball interception task, finding a low-dimensional neural embedding that tracks the ball's position.

The paper shows that low-dimensional neural manifolds enable parametric control of flexible timing through tonic inputs in trained recurrent networks and confirms this in monkey behavior.

This paper introduces AutoLFADS, an automated framework for estimating neural population dynamics from neural data without supervisory information, validated on multiple datasets.

The paper identifies a precise neural mechanism in the frontal cortex of non-human primates that adapts to temporal statistics, supporting predictive processing of time.

This study introduces a method to test whether humans use model-based Bayesian computations by assessing sensitivity to equivalent prior-cost pairs that require different internal models.

The paper shows how prior beliefs influence behavior by shaping cortical latent dynamics through neural activity in monkeys during a time interval reproduction task.