Purpose
The quick CSF method is a novel adaptive method developed to rapidly estimate features of the contrast sensitivity function, including i) peak contrast sensitivity; and ii) CSF acuity, the high-frequency cutoff that corresponds to a contrast threshold of 50% (sensitivity=2). It has been demonstrated that CSF acuity estimates correlate well with logMAR acuity (Lesmes LA, et al. IOVS 2012;52:ARVO E-Abstract 4358). Because acuity is an important criterion for choosing optical correction, we investigated how dioptric blur affects CSF acuity estimates obtained with the quick CSF, implemented on an iPad for convenience.
Methods
Six observers (21-46 yrs old; mean 27.5) were tested at a viewing distance of 40cm. Tests were run at self-reported correction and additional blur conditions (+1,+2,+3,+4D). Gratings of different frequencies (.29-18.5cpd) and contrasts (.2-100%) were presented for 250ms at 3.4deg either left or right of fixation (spatial 2AFC) and subjects used the touch screen to respond. Each test (120 trials) lasted 3-4 min. To estimate repeatability, all conditions were run again the following day.
Results
As expected, dioptric blur primarily affected the CSF's high-frequency region, and not its peak height (Fig 1). When normalized to the lens condition with maximum acuity, CSF acuities exhibited a sharp decrease across lens conditions that was not demonstrated by peak sensitivity (Fig 2). Although contrast thresholds measured on digital displays often exhibit ceiling effects, we reliably obtained peak thresholds <1% for 5 subjects. Test reliability was calculated from the coefficient of variation for difference scores between first and second tests; for peak sensitivity, this was <12% after 30 trials, <6.2% after 60, and <3.8% after 90 and 120 trials. These values for CSF acuity are <16%, <14.4%, <10%, and <7%, respectively.
Conclusions
These results confirm that dioptric blur affects high frequencies more than the peak and lower frequency regions of the CSF. The main contribution of the current study is the use of a portable tablet device, and algorithms to extend contrast resolution (>8bits), to rapidly and reliably measure features of the contrast sensitivity function.
Keywords: 478 contrast sensitivity •
754 visual acuity