Curriculum Vitae   | KIV Model  |  Publications

Mark H. Myers Computational Neurodynamics Laboratory Department of CS, University of Memphis, Memphis, TN 38152 mhmyers@memphis.edu (901) 603-3817

Biologically-Inspired Neural Networks for Modeling Nonlinear Neurodynamics The area of research we focus on is understanding cognitive processing relating to learning, and brain pathologies specifically in the area of epileptic seizures. These areas share a common theme in the concept of the 'electrically balanced brain' . When electrical activity in the brain reaches a particular threshold in large neural group firings, which can be found in seizures, or even 'small' seizure attributes found in the neural firings exhibited during cognitive processing, the electrical activity of the brain attempts to restore itself back to its original electrical state.   Chaotic Attractors found in the Mind We use Nonlinear Neurodynamic tools to decipher and analyze the signals produced through mesoscopic neural groups and captured from EEG data recordings. In order to study the electrical behavior of the areas of the brain that exhibit cognitive processing and seizure behavior, we have developed a dynamic neural network model based on Katchalsky sets ("K sets"). The architecture of the K-sets, specifically the KIV model, exhibit many similar attributes found in biological systems that incorporate the limbic system found the brain architecture of salamanders, primates, and humans. Analysis of the electrical brain signal through signal processing concepts such as power spectral densities and non-linear dynamic tools such as Lyapunov exponents enable us to study the attributes of the brain signal during cognitive processing, awake, sleep, and seizure states. The image above is the inherent chaotic attractors found in a signal that was captured through an EEG.