Personal tools
You are here: Home Events ANC Workshop: Matthias Hennig and Yann Sweeney, Chair: Gabriele Schweikert

ANC Workshop: Matthias Hennig and Yann Sweeney, Chair: Gabriele Schweikert

— filed under:

What
  • ANC Workshop Talk
When Oct 08, 2013
from 11:00 AM to 12:00 PM
Where IF 4.31/4.33
Add event to calendar vCal
iCal

Matthias Hennig

The signal and the noise: Why many events are undetectable in physiological recordings - but some aren't

I will give a summary of our work on neural spike detection in high density multielectrode array recordings. The strategy we have developed to effectively deal with the high data rates in these data is to first perform a fast, online-capable detection of putative spike events, which are then analysed more extensively off-line. To distinguish between signal and noise, we assume that true spikes are likely to show some degree of correlation with other events. This method is shown to yield more detected events than simple a thresholding procedure. Moreover, we found that averaging signals across multiple channels can reduce noise and substantially improve detection of events that would otherwise remain below detection threshold.

 

Yann Sweeney

Can diffusive neurotransmitters provide a homeostatic signal?

Gaseous neurotransmitters such as nitric oxide (NO) provide a unique and often overlooked mechanism for neurons to communicate through diffusion within a network, regardless of synaptic connectivity. Recent experimental studies implicate NO in homeostatic control of neuronal excitability.  We conduct a theoretical investigation of the distinguishing roles of nitric oxide diffusion, or volume transmission, in comparison with canonical homeostatic mechanisms. 

I will attempt to give a brief introduction to homeostasis in the brain and discuss results arising from implementing a simplified model of the above homeostatic mechanism in numerical simulations of a cortical-style network of integrate-and-fire neurons, and in a population of neurons desribed by the mean-field approximation.