Purchase this article with an account.
Arun kumar Krishnan, Mehmet Naci Agaoglu, Susana T L Chung; Assessing fixation stability for 30 seconds – Is it long enough?. Invest. Ophthalmol. Vis. Sci. 2017;58(8):5628. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Fixation stability (FS) is known to be affected by visual and non-visual factors, but little is known about its dependence on test duration. Most studies record gaze for prolonged durations like 30 s. However, to our knowledge, there exists no systematic study that examines how FS depends on the recording duration. The primary goal of this study was to assess the effect of recording duration on FS.
Eleven normally sighted subjects (Age range: 21-35 years) binocularly fixated a high-contrast, black 1° cross displayed against a uniform gray background on a 32” display. Subjects were optically corrected for the distance (76 cm) and eye movements were recorded using an EyeLink 1000 Plus eye tracker. A chin and forehead rest was used. Each test session began with a calibration and validation, and a drift correction was applied at the beginning of each trial. A minimum of 5 trials of 70 s duration were recorded for each subject. We then extracted epochs that ranged between 0.5 and 64 s, in octave-steps, and calculated FS for each epoch. In addition, we also measured fixation for trials of eight different durations ranging between 0.5 and 64 s, in a randomly interleaved sequence. The order of testing fixed-duration (70 s) and mixed-duration trials was counter balanced across subjects. FS was quantified as the area enclosed by the 68% isoline (ISOA) of the distribution of eye positions during fixation.
Across all subjects, ISOA increased (fixation becoming more unstable) with duration. This was observed for both measuring methods, although ISOA determined using the fixed-duration method was in general, slightly smaller than that for the mixed-duration method (by ~14%, p < 0.0001). We fit the data-set of log ISOA vs log duration for each subject and each measuring method with an exponential function, and used 3x the time constant to represent the duration at which ISOA reached 95% of saturation. Across subjects, this duration-to-saturation averaged 99042 s and 8478 s for fixed-duration and mixed-duration methods, respectively.
The duration-to-saturation is much longer than the commonly used 30 s, which makes it not a practical option as a standard duration for measurement. Nevertheless, our finding of the strong dependence of FS on duration implies that comparisons of FS across studies definitely need to take into account the duration of fixation recording, and perhaps the different measurement methods.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.
This PDF is available to Subscribers Only