Normal human contrast sensitivity for static and dynamic sinusoidal gratings. A comparison between two automated...

Contrast thresholds for static and dynamic (phase-shifted at 2 rps) sinusoidal gratings were established for seven spatial frequencies 0.5-32 c/deg. An HP-85 desktop computer ran an automated stimulus descending algorithm. Subjects reported stimulus presence by pressing a switch. Two stimulus presentation designs were used. In method 1, the screen was blanked to mean luminance between presentations, while in method 2 the pattern remained unchanged until the next contrast level was presented. Mean thresholds were calculated from three successive threshold passages using 2 dB steps. Fifty-nine volunteers from a military unit, mean age 20 years, participated in the study. Method 1 always gave lower thresholds with less variability. For static gratings, this was statistically significant at 0.5 and 16 c/deg (p < 0.01 and 0.05, respectively) and for dynamic gratings at all spatial frequencies except 32 c/deg (p < 0.1 for 0.5 and 16 c/deg, p < 0.001 for 1-8 c/deg). An accuracy index was calculated from two determinations at 4 c/deg. Method 1 gave higher indices for both static and dynamic patterns, although not statistically significant. However, method 1 was significantly more time consuming. Method 1 was considered the more reliable of the two contrast threshold determining designs.