Theoretical Neurophysics

Intelligence Modeling

We use a neurophysics research paradigm called “Intelligence Modeling” to convert complex observed behaviors (molecular, cellular and organism levels) into predictive modeling equations that can be parameterized via a variety of measurements. Intelligence modeling is drawn from the intellectual traditions of information theory, systems theory and cybernetics, three closely linked fields of study. Systems theory is concerned with how the organization of the parts of a system give rise to the properties of the system independently of the substrate from which the system itself is composed and therefore how a system interaction and emergent behavior can be predicted from general principles (Von Bertalanffy, 1976). Cybernetics is a term introduced by the mathematician Norbert Weiner and is derived from the Greek for “steersman” because it refers to the whole field of control and communications in the animal and machine (Weiner, 1965). Cybernetics is concerned with the flow of information within a system (such as feedback loops) and with the mechanisms governing how information transmitted can act to control the parts or whole of an organism or system generally leading to both homeostasis and intentional action. Information theory (mostly attributed to Shannon, 1948) is concerned with how information is coded and transmitted accurately between a source and receiver (or between the parts of the system) over transmission channels that are noisy. A key quantity in this treatment is the Shannon entropy (H), a measure of how many bits of information are required for the source to send in order for the receiver to be 100% certain of the state or message of the source. Each bit carries a certain energy cost and the sending of bits whose state are equiprobable are the most energetically costly to send. The capacity to send maximal information at minimal H-cost is a central concern in information theory specifically source coding.

Click here for more on Intelligence Modeling……

Computational and Research Design

Much of our laboratory’s research, both theoretical and at the bench, makes extensive use of computers and computational methods.  We use mathematical modeling and computational chemistry techniques in a number of theoretical lines of research.  Skills in electronics and computational machinery are used almost daily in the laboratory, since the design and adaptation of new and existing instrumentation is necessary in our research efforts. In recent years our laboratory has adapted total internal reflectance fluoroscopy (TIRF) to provide imaging of the membrane potential and closely related cellular regions in a restricted region of spatial interest (approximately 600 nM with a He-Ne laser). TIRF techniques are used to characterize the interphase region in terms of physical constants such as the diffusion coefficients of molecules. These experiments are being developed using TIRF and electrochemical analysis with modified microelectrode arrays in cell culture systems. The instrument designed to study the redox sensitive receptor sites uses a hybrid technology of electrochemical potentiostatic control at a synthetic membrane with fluorescent digital image analysis.