Renormalization Principles in Neural Systems

From Circuits to Cognition
COSYNE 2026 Workshop • Lisbon, Portugal • 16th March 2026

Overview

This full-day workshop explores renormalization as a unifying framework for understanding neural computation across scales: from local circuit dynamics to cognitive representations. We gather leading theorists, experimentalists, and modelers to investigate systematic coarse‑graining in neural systems: from local interactions to global dynamics, across‑level information transformation, and the rise of abstraction and generalization.

Designed for those coming from various backgrounds but similar research goals, the workshop aims to align concepts and methods so attendees leave with a common vocabulary and a shared understanding of the key principles and methods of renormalization in neuroscience.

Key Themes

Multi-Scale Dynamics

Multi-Scale Dynamics

How local interactions generate emergent global behavior across temporal and spatial scales

Higher-Order Structure

Higher-Order Structure

The role of simplicial complexes and topological features in neural computation

Cognitive Representations

Cognitive Representations

Geometric and topological structure of neural representations from cells to systems

Scale Invariance

Scale Invariance

Principles of coarse-graining that preserve essential computational properties

Keynote Speakers

Giuseppe Gava
Giuseppe Gava
University of Oxford

Hippocampal assemblies, scale coding

Pedro Mediano
Pedro Mediano
Imperial College London

Information decomposition & causal emergence

Valeria Fascianelli
Valeria Fascianelli
Columbia University

Computational modeling of behaviorally relevant neural dynamics

SueYeon Chung
SueYeon Chung
Kempner Institute

Representation geometry and abstraction

Keith Hengen
Keith Hengen
Washington University in St. Louis

Homeostatic plasticity and neural criticality across scales

Kenneth D. Harris
Kenneth D. Harris
University College London

Neuropixels recordings & population structure

Ila Fiete
Ila Fiete
MIT

Multi-scale neural representations, grid-cell coding, and the geometry of abstraction

Julijana Gjorgjieva
Julijana Gjorgjieva
TU Munich & Max Planck Institute

Computational neuroscience, synaptic plasticity, circuit development

Workshop Schedule

Organizers

Jesseba Fernando

Jesseba Fernando

Network Science Institute, Northeastern University

Network reorganization during learning, bridging systems neuroscience and AI through information-theoretic approaches.

Giovanni Petri

Giovanni Petri

Network Science Institute, Northeastern University London

Topological data analysis of brain networks, higher-order interactions, and information geometry. Develops frameworks for analyzing multi-scale brain dynamics through algebraic topology.

Andrea Santoro

Andrea Santoro

CENTAI Turin

Higher-order interaction modeling, time-series analysis, and representation geometry in neural systems. Bridges information theory and empirical neuroscience.