Learning effect of Humphrey Matrix perimetry

Background: Humphrey Matrix perimetry is a recent development in automated perimetry that uses frequency-doubling technology (FDT). We evaluated the learning effect of Humphrey Matrix perimetry with a full-threshold 30-2 strategy. Methods: Twenty-four patients with primary open-angle glaucoma and early visual field defects as assessed by standard automated perimetry and 24 healthy control subjects participated. All subjects had no prior experience with FDT visual field tests. Humphrey Matrix perimetry using a full-threshold 30-2 strategy was performed 3 times within 1 month. Various visual field indices were compared across multiple tests. Results: Mean deviation (MD) showed a larger defect at the first test than at the second and third tests (p < 0.001), and the mean total improvement of the MD was 2.80 decibels (dB) in the glaucoma group and 1.71 dB in the control group. Pattern standard deviations (PSDs) were decreased as the tests were repeated, but test duration, fixation losses, and false-positive or false-negative rate were not changed significantly. Additionally, the Glaucoma Hemifield Test (GHT) showed unstable results. Interpretation: In patients naive to the FDT test, Humphrey Matrix perimetry showed a large learning effect on MD and PSD among the various visual field indices. The GHT results of Humphrey Matrix perimetry showed a large instability.