[د/سمير المليجى]

الفيزياء العصبية للمخ



The goal of the theoretical Neurophysics is to uncover fundamental concepts and mechanisms of neural information processing, analyzing, and put into a quantitative connection with empirical observations. Important for us is the acquisition of basic principles behind a plethora of seemingly independent individual observations. Our department attaches great importance to working closely with experimental or application specific working groups for all projects. The emphasis in our cooperation on the one hand in the development and transfer of new methods of data analysis. In parallel, models and mechanisms of neural information processing by means of targeted experiments on their explanatory power. Theoretical studies are complemented by application-oriented projects in which concepts of neural processing are transmitted to the problems of robotics and image analysis.

In current projects we are interested in the following questions: What are the spatio-temporal states in the brain are responsible for the perception of objects? How do these states, and on what time scale they exist? In what way information is learned and stored? What code uses the brain to edit this information? What are the legalities signals from the environment with states are combined in the brain to generate meaningful behavior in a dynamic environment?

Our hypothesis is that the relevant object recognition and information processing behavior arises from the interaction and precise coordination of individual neurons, which manifests itself in spatio-temporal patterns of activity and interconnection. To this end, we investigate biologically realistic models of neural processing that are numerically analyzed with the help of large computer simulations and mathematical methods of nonlinear dynamics and statistical physics. Current projects focus on, for example, with the basics of contour integration as a first step, an object-based image analysis. Other projects explore the coding of individual and combined features of visual stimuli. In addition to an information-theoretic treatment of the basics of this question, we are particularly interested in the role of attention during encoding and perception of objects. One focus of applied research in our group is the development of optimized, robust algorithms for detecting typical brain patterns that can be used in Brain-computer interfaces for control of prostheses. Equally important for the understanding of the brain and as an application in robotics is the analysis of learning and behavioral algorithms for orientation of autonomous agents in a dynamic environment. Theoretical concepts are available here translated into practice and thus lead to a deeper understanding of the emergence of behavior through interaction with an environment.