Fig. 5
From: Extracting single-trial neural interaction using latent dynamical systems model
Analysis of optimal models via logistic regressions including DONA and spike counts \({N}_{sp}\). a Percentage of neurons for which spike counts or DONAs were significantly associated with the probability of the lick response (p < 0.05 in the logistic regression). The dotted line represents the significance level (binomial test, p < 0.05). b Percentage of neurons for which the optimal model (smaller AIC) in the logistic regression was the model including only spike counts and the full model including spike counts and DONA. c, d Example dynamics of the spike counts and DONAs of a neuron for which the probability of the lick response depends significantly on both variables of the neuron in the delay epoch. Solid and open circles represent the mean values of left and right lick responses, respectively, whereas shaded regions denote the standard error of the mean (SEM)