Abstract

Recovery of the human ERG a-wave can be tracked using a paired-flash protocol, in which the test flash is followed at varying times by a bright probe flash that saturates the rods. Birch et al.¹ used this approach to measure the period (T) of apparent saturation of the rod-mediated a-wave² in relation to the intensity (I_f) of the test flash. Fig. 1 shows the "saturation function" (T vs. ln I_f function) obtained in a group of experiments. Over the investigated range of I_f (40 to 10⁶ scot-td-s), the saturation function exhibits two essentially linear branches, the slopes of which [ΔT/Δ(ln I_f)] are ≃0.3 s and ≃2.3 s. Results obtained in studies of salamander rods^3-5 led Birch et al. to hypothesize that the upper-branch slope of 2.3 s represents "τ_R*", the exponential lifetime of photoactivated rhodopsin (R*). The occurrence of two branches in the saturation function of Fig. 1 could in principle reflect a variation of τ_R* with flash intensity, i.e., an increase of τ_R* from -0.3 s to =2.3 s with increasing I_f. However, the evidence that τ_R* in salamander rods is unaffected by changing illumination conditions^3-6 suggests that τ_R* in human rods may be constant. Here we present a model of rod transduction based on the assumption that τ_R* is constant. We show that this model can account for the two-branched property of the a-wave saturation function.

© 1996 Optical Society of America