Related Experiment Video
Updated: May 15, 2026
16:01
Multi-electrode Array Recordings of Neuronal Avalanches in Organotypic Cultures
Published on: August 1, 2011
Asymmetry in neural fields: a spatiotemporal encoding mechanism.
Mauricio Cerda1, Bernard Girau
1Laboratory for Scientific Image Analysis SCIAN-LAB at the Program of Anatomy and Developmental Biology and the Biomedical Neuroscience Institute BNI, ICBM, Faculty of Medicine, Universidad de Chile, Independencia, 1027 Santiago, Chile. mauriciocerda@med.uchile.cl
Biological Cybernetics
|January 9, 2013
Summary
Neural field models with asymmetric connectivity bias neural population activity. This study shows asymmetric neural fields offer competitive spatiotemporal encoding for online classification and distributed operation.
Area of Science:
- Computational Neuroscience
- Theoretical Neuroscience
Background:
- Neural field models are established tools for understanding brain functions like attention and memory.
- Research has predominantly focused on symmetric neural connectivity, overlooking asymmetric dynamics observed in cortical tissue.
Purpose of the Study:
- To investigate the emergent properties of neural fields with asymmetric connectivity.
- To analyze how asymmetric connectivity influences neural population dynamics, specifically front propagation.
Main Methods:
- Exploration of neural field dynamics under asymmetric connectivity conditions.
- Analysis of front propagation and activation trajectories.
- Encoding of human motion video sequences using asymmetric neural fields.
Main Results:
- Asymmetric connectivity biases neural populations towards specific activation trajectories.
- A linear relationship was identified between asymmetry and input speed for localized inputs.
- The asymmetric neural field demonstrated competitive performance against computer vision techniques for video encoding.
Conclusions:
- Asymmetric neural fields present a viable alternative for spatiotemporal encoding.
- Key advantages include suitability for online classification and distributed processing capabilities.
