Distribution <i>P</i>(<i>r</i>∣<i>s</i>) of network responses to a stimulus <i>s</i>.

<p>Normalized histogram of 10⁵ averages of <i>n</i> random individual neuron firing rates (symbols), and <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0121794#pone.0121794.e054" target="_blank">Eq (30)</a> (lines). For evaluation of <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0121794#pone.0121794.e054" target="_blank">Eq (30)</a> we calculated the mean stationary network activity </p><p></p><p></p><p></p><p><mi>r</mi><mo>^</mo></p><mo stretchy="false">(</mo><mi>s</mi><mo stretchy="false">)</mo><p></p><p></p><p></p> using the self-consistent <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0121794#pone.0121794.e010" target="_blank">Eq (5)</a>. The corresponding variance of the sampling noise <i>σ</i>(<i>s</i>)²/<i>n</i> was obtained from the approximate stationary firing rate distribution as previously discussed for <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0121794#pone.0121794.g003" target="_blank">Fig 3B</a>. The distributions are broadened by assortativity, corresponding to larger noise of the response of assortative networks.<p>