Skip to main content


Fig. 6 | BMC Neuroscience

Fig. 6

From: Empirical Bayesian significance measure of neuronal spike response

Fig. 6

Results of Experiment 4 that assessed functional connectivity detection performances at various spike frequency and connectivity weights with fixing time sample number at \(T=2000\). Spike frequencies and connectivity weights are controlled by parameters A and w, respectively. Rows and columns of the 36 panels denote variations of A and w values, respectively. In each panel, we compared four ROC curves; red and blue line color denote estimation methods of GLM, L2 regularization and variational Bayes estimation, respectively. Real and dotted line styles denote Peak and MaxZ statistics, respectively. The combination of GLM with L2 regularization and Peak statistic performed the best for most of the 36 panels

Back to article page