Mechanisms mediating visual detection in static perimetry

Abstract

Purpose: The usual stimuli in static perimetry are white-light luminance increments. However, the specific visual detection mechanisms involved in perimetry are unknown because all classes of neural mechanisms are sensitive to spectrally broadband stimuli. The objective of this study was to determine the relative sensitivities of nonopponent and opponent detecting mechanisms under standard perimetry test conditions.

Methods: Using trained rhesus monkey subjects, the relative sensitivities of the vision mechanisms for the detection of perimetry test stimuli were determined through psychophysical measurements of spectral sensitivity at each of the test field locations of the C24-2 threshold program on the Humphrey Field Analyzer (Allergan Humphrey, San Leandro, CA). The spectral sensitivity functions were analyzed by a three-channel model that incorporated independent short-wavelength-sensitive, nonopponent (luminance), and opponent (chromatic) spectral sensitivity mechanisms.

Results: The visual detection mechanisms for perimetry thresholds varied as a function of the size and wavelength of the test field. With the perimeter's standard stimulus (Goldmann Size III) and bowl illumination (31.5 asb), the presence of a short-wavelength-sensitive mechanism was clearly evident at all field locations, but its relative sensitivity systematically declined with eccentricity. Under these conditions, the sensitivities of the opponent and nonopponent mechanisms were approximately equal at most field locations. With a larger stimulus (Goldmann Size V), however, the contribution of the opponent spectral sensitivity mechanism was more apparent over most of the central field and the alterations of sensitivity with eccentricity were less pronounced. In contrast, a small test field (Goldmann Size II) appeared to bias detection toward nonopponent mechanisms.

Conclusion: The results of these investigations indicate that detection thresholds during perimetry can be effectively biased toward different photopic, visual processing channels through the appropriate selection of size and wavelength of the test stimulus.