The standard oculokinetic perimetry test (OKP) was modified to present a light blue stimulus on a dark background (MOKP) to determine whether the sensitivity and specificity for detecting glaucomatous visual field loss could be improved. Thirty-five adult glaucoma patients (70 eyes) self-administered both tests and the results were correlated with the loss of retinal sensitivity on the Octopus IV program 38. The MOKP detected 18% more true scotomatous loci (more than 15 dB loss of attenuation) than the standard OKP (p < 0.0001). The gain was due to 37% fewer false negatives (p < 0.0001), but with a doubling of the false positives rate (p < 0.049). With a disease prevalence of approximately 2%, the MOKP and OKP would respectively miss 15% or 30%, and include 14 or 9 normals for each diseased individual. This relatively simple modification may further improve the OKP for detecting glaucoma.