Berlin 2015 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
BP: Fachverband Biologische Physik
BP 14: Posters: Neurophysics
BP 14.7: Poster
Monday, March 16, 2015, 17:30–19:30, Poster A
Can compartmentalization explain fast population coding? — •David Hofmann1,2, Andreas Neef1,2, and Fred Wolf1,2 — 1Max Planck Institut für Dynamik und Selbstorganisation, Göttingen — 2Bernstein Center for Computational Neuroscience, Göttingen
Cortical neurons, driven by noisy current injections, change their firing rate within 1 ms of a small step in the current average. This sets the speed of cortical information processing. Theoretical and experimental evidence points at the rapidness of the action potential (AP) onset as the key determinant of the fast neuronal response. However, the biophysical basis of this onset rapidness is unclear and a matter of current debate.
Recently, a minimal multi-compartment model was presented that produces a rapid onset in the somatic AP waveform (Brette PloS Comp. Neuro. 2013). This occurs by electric decoupling of the soma from the site of AP initiation (compartmentalization).
Here we investigate whether the electric decoupling mechanism can also explain the fast neural response. Specifically, we tested whether the model reproduces two robust experimental observations, a) the high cut-off frequency of the dynamic gain function in the range of 200 Hz b) the increasing cut-off frequency for increasing input current correlation time which is called the Brunel effect.
We find that the gain function is dominated by a single pole low pass filter around the membrane time constant independent of the electric decoupling. Hence, the model is not able to explain fast population responses. In addition, the model does not display a Brunel effect.