Comparing Different Measures of Spatio-Temporal Patterns in Neural Activity
Advances in the technology of multi-unit recordings have created the need for new statistical approaches to detect the presence of spatiotemporal patterns in neuron spike train data. We present statistical approaches to detect synchronisation in the activity of three or more neurons. These phenomena must be modelled as higher order correlations that cannot be reduced to a simple combination of second order correlations. We examine three measures for the presence of higher-order patterns of neural activation: coefficients of log-linear models, connected cumulants and redundancies. We present arguments in favour of the coefficients of log-linear models. We introduce the Constraint-Perturbation-Procedure (CPP) as an alternative to Iterative proportional Fitting to construct test statistics for detecting the presence of higher order interactions. The methods are applied to experimental data drawn from of multi-unit recordings from the frontal cortex of behaving monkeys.
KeywordsHigh Order Interaction High Order Correlation Binary Neuron Binary Configuration Iterative Proportional Fitting Procedure
Unable to display preview. Download preview PDF.
- Hebb, D. (1949) The Organization of Behavior. New York: Wiley, 1949.Google Scholar