ABSTRACT
We propose a new analytical method for determining the response threshold in electroretinogram (ERG) in which the wave shows a biphasic slow dc-potential shift. This method uses the recorded wave to the highest intensity stimuli in each wavelength tested as a template wave f(t), and it was compared with other recorded waves obtained under lower intensities g(t). Our test recordings in medaka Oryzias latipes were analogous between the template and the compared waveforms, although there were differences in amplitude and time lag (τ, peak time difference) which occurred as a result of the difference in stimulus intensity. Cross-correlation analysis was applied. Based on the obtained cross-correlation function Cfg(τ) in each comparison, τ was determined as the time lag at which the cross-correlation coefficient Rfg(τ) showed the maximum value. Determined thresholds that were based on both the experimenter's visual inspection and this new method agreed well when the adoption condition was set to satisfy R(τ) ≥ 0.7 and τ ≤ 150 ms in scotopic or τ ≤ 120 ms in photopic conditions. We concluded that this "template wave matching method" is a quick and reliable objective assessment that can be used to determine the threshold. This study analyzed ERG recordings in response to 6 kinds of wavelength light stimuli (380 nm to 620 nm) at different photon flux densities. We report the threshold levels and relative spectral sensitivities in scotopic and photopic vision of medaka.
