The Dynamical Systems seminar is held on Monday afternoons at
4:00 PM in CCDS 365 . There will be a brief tea in CCDS 365 at 3:45PM.
January 26th Greg Handy (U of Minnesota - Twin Cities) - TALK CANCELLED DUE TO BU SNOW DAY, we plan to reschedule later in the semester.
Title: Glial ensheathment of inhibitory synapses drives hyperactivity and increases correlations
Abstract: Recent evidence shows that glial cells actively modulate neuronal dynamics. A recent study notably found that during and after anesthesia, microglia ensheath inhibitory synapses, disrupting neurotransmitter flow. In this talk, I will develop computational models that explore how this ensheathment affects neuronal dynamics. Extending a microscale synaptic cleft model, I show that ensheathment accelerates synaptic transmission but reduces its strength. I will then integrate this microscale model into a large network of exponential integrate-and-fire neurons, which introduces heterogeneous synaptic parameters determined by glial proximity, and extend linear response theory to analyze firing rates and noise correlations. I will show that this model reproduces the experimental finding that increased glial ensheathment of inhibitory synapses leads to hyperactivity and predicts significant increases in power spectrum magnitude across task-relevant frequencies, suggesting glial-driven synaptic plasticity is an underappreciated mechanism for modulating cortical dynamics.
February 2nd Petur Bryde (MIT)
Title: Orientational dynamics and control of topological defects in 2D nematics
>
Abstract:
Topological defects play a distinguished role in the dynamics of ordered media and present promising candidates for robust information storage and computation in soft matter systems. In particular, half-integer defects in nematic films carry orientational degrees of freedom whose long-range interactions can be used to realize nematic analogs of classical logic gates as well as generalized continuous logic functions. Designing such “nematic circuits” with prescribed input-output behavior requires a reduced-order model for the orientational dynamics of a system of defects which is amenable to control-theoretic methods. Here we introduce a framework in which the nematic director angle, which satisfies a linear diffusion equation, is coupled to the defect orientations via appropriate boundary conditions around the defect cores. This approach yields an analogy to heat conduction and enables the design of control protocols that can be implemented in liquid crystals using existing experimental techniques.