Analysis of Parallel Spike Trains

Volume 7 of the series Springer Series in Computational Neuroscience pp 191-220

Unitary Event Analysis

  • Sonja GrünAffiliated withLaboratory for Statistical Neuroscience, RIKEN Brain Science Institute Email author 
  • , Markus DiesmannAffiliated withLaboratory for Computational Neurophysics, RIKEN Brain Science Institute
  • , Ad AertsenAffiliated withNeurobiology and Biophysics, Institute of Biology III, Faculty of Biology, Albert-Ludwig University

* Final gross prices may vary according to local VAT.

Get Access


It has been proposed that cortical neurons organize dynamically into functional groups (“cell assemblies”) by the temporal structure of their joint spiking activity. The Unitary Events analysis method detects conspicuous patterns of coincident spike activity among simultaneously recorded single neurons. The statistical significance of a pattern is evaluated by comparing the number of occurrences to the number expected on the basis of the firing rates of the neurons. Key elements of the method are the proper formulation of the null hypothesis and the derivation of the corresponding count distribution of coincidences used in the significance test. Performing the analysis in a sliding window manner results in a time-resolved measure of significant spike synchrony. In this chapter we review the basic components of UE analysis and explore its dependencies on parameters like the allowed temporal imprecision and features of the data like firing rate and coincidence rate. Violations of the assumptions of stationarity of the firing rate within the analysis window and Poisson statistics can be tolerated to a reasonable degree without inducing false positives. We conclude that the UE method is robust already in its basic form. Still, it is preferable to use coincidence distributions for the significance test that are well adapted to particular features of the data. The chapter presents practical advice and solutions based on surrogates.