Fig. 7
From: Propagation of temporal and rate signals in cultured multilayer networks
Roles of AMPA and NMDA-mediated components in the propagation of activity. a Example of dot rasters of spikes recorded from individual neurons in each layer (layers 1–4 from black to red) in cell-attached mode in response to synchronized activation of neurons in the first layer (30 repetitions) under control conditions (left), in the presence of 50 µM APV, and in the presence of APV and 10 µM bicuculline (right). b Average poststimulus time histogram in each layer. Data were compiled from 9 networks of density = 278 ± 68 neurons·mm−2 (mean ± STD). c Left, average firing rate half width vs layer (mean ± SEM). Right, data combined across layers shown as whisker plots (box: median and interquartile range, whiskers: full range of the distribution; outliers are plotted individually). d Similar plots for the firing rate. Mann–Whitney U-test statistics for the width: control vs APV: P = 7.5 × 10−6, APV vs Bicuculine: P = 2.7 × 10−6, control vs Bicuculine: P = 0.85. Mann–Whitney U-test statistics for the firing rate: control vs APV: P = 3.2 × 10−4, APV vs Bicuculine: P = 1.1 × 10−7, control vs Bicuculine: P = 0.008. ns not significant, *P < 0.05, **P < 0.01, ***P < 0.001. e Channel capacity quantified as the probability of correctly classifying firing rate in the first layer given the firing rate in the nth layer. All data are shown and the mean ± SEM is represented as thick lines. Mann–Whitney U-test statistic for the comparison of decoding accuracy of neurons in layers 2–4 vs conditions: control vs APV: P = 0.010, APV vs Bicuculine: P = 0.032, control vs Bicuculine: P = 0.954
