Learning effect, short-term fluctuation, and long-term fluctuation in frequency doubling technique.

Purpose: To evaluate short-term and long-term fluctuations and learning effects in healthy subjects tested with the frequency doubling technique, which is based on a low spatial frequency and a high temporal frequency stimulus.

Methods: Twenty-three healthy adult subjects were consecutively recruited from volunteers, and 20 subjects finished the study. All the visual fields were assessed by the frequency doubling technique, program C-20, full threshold. The frequency doubling technique presents stimuli on a black-and-white video monitor with specialized control circuitry interfaced to a microprocessor. During program C-20, full threshold, 17 points are tested, one round point centrally and 16 square ones in the periphery up to 20 degrees of eccentricity. Each stimulus consists of a 0.25-cycle/degree sinusoidal grating undergoing a 25-Hz counterphase flicker. One eye of each patient was chosen at random. Each subject was tested once in the first session, three times in the second session, and once in the third and fourth sessions. Both short-term and long-term fluctuation were studied either as the average fluctuation value of all the points tested or as a point-to-point fluctuation. To study the learning effect, the results of the first session were compared with those of the second, the third, and the fourth sessions.

Results: The average mean sensitivity of the three examinations of the second session was 30.4 +/- 1.24 dB, and the average short-term fluctuation of all the subjects was 2.16 +/- 0.5 dB. The short-term fluctuation of each point tested ranged from 1.4 to 3.4 dB. The average mean sensitivity for all the sessions was 32.4 +/- 1.14 dB, with an average long-term fluctuation of 3.23 +/- 0.5 dB, and the long-term fluctuation of each tested point ranged from 2.5 to 4.4 dB. A mild learning effect was found between the first section and the others.

Conclusion: Short-term and long-term fluctuations were similar to those known to occur with the conventional threshold perimetry when they were compared with the literature data. A learning effect was also observed and should be taken into account for the clinical use of this test.