Statistical and dynamical criticality of human brain ECoG activity

G SOLOVEY

Seminario; Seminar on computational biology, IBM Watson Research Center; 2010

A salient property of the brain is its ability to change due to sensory inputs (plasticity) while being stable under perturbations (robustness). However, dynamical systems theory, as well as the theory of attractor networks that has dominated the field of neural computation, both hold or assume that physical systems should be structurally stable. In this talk I will present evidence that human brain ECoG recordings are close to the onset of instability (structurally unstable) and display features of both dynamical and statistical criticality.