[Comp-neuro] Re: Similarity Measures for Spike Trains

Naud Richard richard.naud at epfl.ch
Tue Dec 6 11:04:11 CET 2011


Hi Harry,

I have read that article too!  If I understand your question, you would like to know how to quantify the distance between spike trains generated by a process that can stochastically switch between, say, two distinct spiking mechanisms (like up/down or bursting/regular).  But what is the important quantity? How the modes are mixed or how each mode is characterized.  I'd agree that the problem seems ill posed, and I don't know of anyone addressing the issue beyond the work of JM Fellous and PHE Tiesinga, or the literature on the estimation of hidden Markov models.

Cheers,

Richard Naud


More information about the Comp-neuro mailing list