Back to the Future: Anticipatory synchronisation as a core mechanism for interarea brain communication

Introduction

Our brain processes multitudes of signals based on recurrent network activity involving feedforward and feedback interarea communication. To funnel information efficiently between areas, low-frequency rhythmic neural activity patterns couple to each other.

Problem Statement

It is an unresolved question how early sensory cortex (SC) can couple efficiently to both feedforward sensory information flow and feedback information flow arriving from higher order regions such as prefrontal cortex (PFC). Since PFC and sensory input lack direct coupling and operate with their own temporal dynamics, SC decouples from the sensory input when PFC couples to SC.

Current Limitations

It is therefore not possible to achieve efficient multi-area coupling in currently proposed feedback-driven coupling schemes.

Proposed Mechanism

In search of a mechanism to address this problem, I resort to a dynamical systems process called anticipated synchronization (AS). During AS, a receiving system (PFC) sends a copy of its own activity as delayed feedback to itself, allowing this system to - paradoxically - anticipate the dynamics of the driving system (SC).

Process of Anticipated Synchronization

During AS, information from PFC is sent back to SC arriving at the right time in the future, while SC maintains sensory coupling.

Research Goals

My goal in the next 5 years is to provide evidence of interarea AS coupling by:

1. Creating a biophysical model of AS.
2. Empirically testing its predictions using ECoG data, psychophysics, and EEG.
3. Validating the use of tACS to control coupling.

Experimental Approach

I will experimentally induce AS in a new closed-loop set-up that stimulates PFC with its own delayed activity.

Expected Outcomes

The proposed research will settle a critical theoretical debate on temporal coordination of interarea brain communication and will provide the computational basis for the hitherto unexplained excess of local feedback loops in the brain.

Implications

The findings will inform new closed-loop treatment approaches and can improve biologically-inspired artificial intelligent systems that currently disregard the exact timing of information flow.