I plotted SNRmax against the sum SNRexc?+?SNRinh and found that the resulting data points lined up close to the diagonal (Fig.?3). noise have not distinguished between ON and OFF pathways. Here I present evidence of marked asymmetries between users of a functional pair of sustained ganglion cells in the mouse retina. The ON cell exhibited a proportionately greater loss of signal\to\noise power ratio (SNR) from its presynaptic arrays to its postsynaptic currents. Thus the ON cell combines transmission and noise from its presynaptic arrays of bipolar and amacrine cells less efficiently than the OFF cell does. Yet the inefficiency of the ON cell is usually compensated by its presynaptic arrays providing a higher SNR than the arrays presynaptic to the OFF cell, apparently to improve visual processing of positive contrasts. Dynamic clamp experiments were performed that launched synaptic conductances into ON and OFF cells. When the amacrine\modulated conductance was removed, the ON cell’s spike train exhibited an increase in SNR. The OFF cell, however, showed the opposite effect of removing amacrine input, which was a decrease in SNR. Thus ON and OFF cells have different modes of synaptic integration with direct effects around the SNR of AP1867 the spike output. and plots, the membrane potentials were corrected after the recording for the liquid junction potential (inh )?inh ? +?(exc )?exc ?,? (1) where: ?inh inh exc exc SNR NMDA plots were fitted with eqn (3) substituted with eqns (1), (2) and?(4) (LevenbergCMarquardt algorithm). The freely fitted parameters were the changes in excitatory and inhibitory conductances and their standard deviations (inh , exc ). Except where noted, the constrained parameters were the reversal potentials AP1867 of these conductances set by the pipette solution for voltage clamp (plots from the same data resulted in unambiguous values for conductances. For the ON cells, fitting plots indicated a small positive inhibitory conductance, and larger positive excitatory conductance (200C500?m diameter spot, ?100 to 100% contrast, was calculated as plots were fitted by eqn (3) equally well whether the Hbb-bh1 voltage dependency of test, and and?and?and?and?and and plots). Plots are 3\point box smoothed and fitted by eqn (3). The reversal potentials for excitatory or inhibitory AP1867 conductances under voltage clamp are shown by vertical dashed lines (and plots. The error bars show standard error of the mean. Spots were of 100C500?m diameter, ?100 to 100% contrast. [Color figure can be viewed at wileyonlinelibrary.com] Plots of SNR against membrane potential were constructed (Fig.?1 and?plot that smoothly transitioned between voltages. To fit these curves, the standard equation used to fit curves was modified to include parameters for light\evoked conductances and their standard deviations over stimulus repetitions (eqn (3) substituted with eqns (1), (2) and (4), Methods). Directly from SNR/curves, it was possible to read off the SNR from each presynaptic array. SNRexc from the bipolar array was the intersection of the curve with the reversal potential for inhibitory conductances (chloride, and?curves, it was possible to read off peak SNR (SNRmax), which was 165??32 for ON cells and 52??10 for OFF cells. Therefore, comparing ON and OFF cells, the ON cell’s presynaptic arrays provided an approximately 3\fold greater SNR than the OFF cell’s arrays did. Estimates of SNR are not distorted by space clamp Compartmental modelling of retinal ganglion cells indicates that the voltage at the soma required to reverse inhibitory or excitatory currents from the dendrites can be different from plots were fitted with eqn (3), which gave values for curves (measurements plot is fitted with eqn (3) (ON sustained cell, Methods). One value of plot best (plot in control condition. plot with inhibitory conductances blocked (same cell). The intercept of the fitted curve with zero current (dashed horizontal line) is the estimated value for plots using the expected values for plot, a conventional plot may not warn of space clamp failure. AP1867 This is because, when fitting an plot, an error in plot, the parameters are not ambiguous in this way (see Fig.?2). For example, early in a previous study (Homann & Freed, 2017), some SNR/plots were poorly fitted with eqn (3) using expected values of curve had a clear singular maximum, SNRmax, that occurred either close to the chloride potential or slightly depolarized to it (Fig.?1). It can be shown, by taking the derivative of the equation that was used to fit SNR/plots, that the maximum value of this equation approximates the sum of SNRs from bipolar and amacrine cell arrays (SNRexc?+?SNRinh) (Methods.