Category: State-space models

• Using a particle filter to decode place cells

  In the last post, I discussed using an extended Kalman filter to decode place cells, based on the algorithm published in Brown et al. (1998). The results looked pretty good. EKFs are certainly better than population vector approaches that don’t consider the sequential nature of the decoding task. The fact that the path of the…

  February 7, 2012

• Using an iterated extended Kalman filter to decode place cells

  Decoding neuronal activity is a powerful technique to study how information is encoded in a population and how it might be extracted by other brains areas. Hippocampal place cells are a prime example of a system that can be studied fruitfully from a decoding persepective. In a typical place cell decoding experiment, a population of…

  February 6, 2012

• GLMs and Hidden Markov models for single neurons

  I posted recently about modeling neurons with continuous state-space dynamics. It’s also possible to model neurons with Hidden Markov models (HMMs), which are state-space models with discrete rather than continuous states. In this post I’m going to focus on the application of HMMs to single neuron data. Single neurons with simple states Suppose that a…

  October 21, 2011

• State-space GLMs for neural data: a simple example

  State space methods are a very broad class of methods to analyze temporal data. At its most basic, you assume that your temporal data is generated by a model with underlying parameters. The parameters are time-varying, and usually it’s assumed that the temporal variation is generated through Markovian dynamics. That is, at every time-step, the…

  October 10, 2011