Skip to content
Yoshihito Saito Ph.D.
Research Background

Research

How does the brain implement intelligence?My research addresses this question by integrating reverse and forward engineering across multiple scales,iterating between experimental measurement and theoretical modeling to build a unified understanding of how neural circuits give rise to adaptive computation.

Ongoing Research Projects

1. Neural circuit mechanisms underlying structure learning and flexible decision-making

Large-scale neural activity is recorded across cortical and hippocampal regions using high-density silicon probes while animals learn the latent structure of their environment. These empirical datasets are integrated with reinforcement learning frameworks and theoretical modeling to uncover how distributed circuits coordinate adaptive behavior.

Conducted at the Brain Computation & Behavior Lab (PIs: A. Fernández-Ruiz & A. Oliva, Cornell University)

2. Mechanistic network modeling of cortical-hippocampal dynamics and functions

Biologically constrained network models are constructed to computationally reproduce cortical-hippocampal interactions. By implementing realistic circuit architectures and learning rules, these models synthesize empirical dynamics to reveal how inter-regional coordination gives rise to neural representations.

In collaboration with T. Asabuki, Ph.D. (RIKEN CBS)

3. Network dynamics of cortex-wide fast-timescale interactions

Mesoscale voltage imaging during cognitive tasks captures cortex-wide neural activity at millisecond resolution. Applying dynamical systems modeling to these extensive datasets extracts the fundamental rules of how rapid, coordinated interactions across distributed cortical networks orchestrate macroscopic information processing.

In collaboration with M. Murayama, Ph.D. (RIKEN CBS)

Past Research Projects

Amygdalo-Cortical Dialogue for Memory Enhancement by Emotional Association

Investigated how coordinated neural activity between the amygdala and distributed cortical networks facilitates memory enhancement through emotional association.

Published in Neuron (2025)DOI: 10.1016/j.neuron.2025.01.001

Large-Scale Analysis of Multi-Regional Cortical Network Dynamics

Analyzed wide-field two-photon microscopy data capturing concurrent activity from over 10,000 neurons across 14 cortical areas, developing an efficient computational pipeline to systematically investigate macroscopic inter-regional interactions and global network properties.

Key Applications

Systems-Level Interrogation of Gut-Brain Interactions

Explored ascending gut-to-brain signaling pathways to broadly understand how gastrointestinal inputs influence the central nervous system and modulate global brain states.

Presented at: The 46th Annual Meeting of the Japan Neuroscience Society (2023)

In Vitro Reprogramming of Astrocytes into Neural Stem Cells

Investigated the specific culture conditions required to reprogram mature astrocytes back into neural stem cells, elucidating the underlying signal transduction mechanisms driving this dedifferentiation process.

Presented at: The 60th Annual Meeting of the Japan Society of Cerebral Blood Flow and Metabolism (2017)