Because of a bug in simulation code, the size of the NMDA conductance evoked by excitatory synaptic inputs was 0.65 nS*msec at threshold instead of 7.2 nS*msec as specified in the Methods section. of the article “Multiplicative Gain Changes Are Induced by Excitation or Inhibition Alone,” by Brendan K. Murphy and Kenneth D. Miller, which appeared on pages 10040-10051 of the November 5, 2003 issue. Therefore the NMDA/AMPA ratio of time integrated conductance at threshold voltage was 0.23 instead of 2.57 as stated. Correcting this by making the NMDA conductance 7.2 nS*msec results in firing rates that are much too high. Simulations with the correct ratio of AMPA and NMDA conductances, but with amplitudes set to produce similar firing rates as in the paper (AMPA 0.91 nS*msec, NMDA 2.34 nS*msec at threshold), produced results essentially identical to those in the paper.