During the 1980s, there was an explosion of information about the physiology of single retinal cells. Although electroretinogram (ERG) research became more sophisticated with modern recording and computer technology, the analysis of the human ERG did not keep pace with the new information that was gained from retinal physiologists. In 1989, we started our study of the cellular bases of the human ERG with a simple strategy. We asked, “Would a quantitative comparison between single-cell physiology and aspects of the ERG help to reveal the cellular bases of the human ERG?” In particular, the analyses of the human ERG at that time did not make use of the new information about primate receptors. In fact, the prevailing analysis of the ERG was actually inconsistent with the single-cell results.¹⁹

Our starting point was the model of the ERG proposed by Granit in the 1930s.^15,16 The vertebrate ERG potential shows two prominent peaks: the a- and b-waves (see chapters 15 and 26). These waves result from the algebraic sum of a number of components with different cellular bases (see chapter 15). Figure 35.1A (from Hood and Birch²²) presents a simplified version of the ERG based on Granit's classic analysis.^15,16 According to this view, the ERG is the result of the algebraic summation of the negative potential P3 produced by the receptors with the positive potential P2 produced by the cells of the inner nuclear layer.

Figure 35.1.  

A, A simplified version of Granit's model^15,16 of the vertebrate rod ERG. (From Hood DC, Birch DG: Human cone receptor activity: the leading edge of the a-wave and models of receptor activity. Vis Neurosci 1993; 10:857–871.) B, The ERGs elicited by flashes of intensities higher than are typically used in the clinic. (Modified from Hood DC, Birch DG: The relationship between models of receptor activity and the a-wave of the human ERG. Cl Vis Sci 1990; 5:293–297.)

We asked, “Is the leading edge of the a-wave the sum of rod receptor activity?” While it was clear that the a-wave largely reflected photoreceptor activity,^7,42 other evidence suggested that postreceptoral contributions could influence the a-wave.^6,40 We argued that if the leading edge of the a-wave is the sum of rod receptor activity, then it should be possible to fit the leading edge of the human rod a-wave with the quantitative model that Baylor et al.² fitted to the responses of single primate rod photoreceptors. In 1990, we showed that the same model, fitted to recordings from single rods, fit the leading edge of the human a-wave.^20,21 The solid curves in figure 35.1B are human rod ERGs to a series of high-intensity flashes presented to the dark-adapted eye. The dashed curves are the predictions from a version of the Baylor et al.² model. These results provided strong support for Granit's view that the leading edge of the a-wave was the response of the receptors.

In this chapter, we review how models of phototransduction have been fitted to the leading edge of the human a-wave and illustrate how the parameters of these models provide a way to evaluate the effects of retinal disease on the human photoreceptors.