Assunta Di Vizio, Riccardo Rolandi, Alessandro Duse, Giulia Carlotta Rizzo, Roberta Daini, Marialuisa Martelli, Marco Barbieri, Gabriele Vestri, Silvia Tavazzi, Fabrizio Zeri
CSF outcomes depend on stimulus type when measured on the same investigated LCD system. Grating stimuli are susceptible to ceiling effects, whereas letters demonstrate good reliability and provide clinically useful normative data, supporting their use in routine functional vision assessment.
PURPOSE: To compare contrast sensitivity function (CSF) measurements obtained with either Gabor gratings or Sloan letters using the same LCD-based device, to evaluate the reliability of letter-based CSF testing and to establish normative values in a young healthy population.
METHODS: A new computerised chart for psychophysical measurements in clinical settings was used (Vision Chart, CSO). Three related investigations were conducted. In Study 1 (n = 44), a randomised crossover design measured monocular CSF with gratings and letters at four spatial frequencies (1.5, 3.0, 6.0 and 12.0 cycles/degree) in dominant and non-dominant eyes. In Study 2 (n = 31), letter-based CSF was measured in the dominant eye across three consecutive trials and reassessed in a second session to evaluate intra- and intersession reliability. A two-way repeated-measures ANOVA evaluated stimulus type and spatial frequency effects in Study 1. Reliability in Study 2 was assessed using repeatability and agreement statistics. In Study 3, normative data were derived by pooling findings from Studies 1 and 2 (dominant eyes n = 75; non-dominant eyes n = 44).
RESULTS: In Study 1, CSF measurements were significantly higher for gratings than for letters at all spatial frequencies tested in both eyes (Holm-corrected p < 0.001). Significant main effects of stimulus type and spatial frequency, as well as a significant interaction, were observed (all p < 0.001). Grating measurements clustered near the upper measurable limit, indicating a ceiling effect. In Study 2, letter-based CSF showed good within-session reliability (intraclass correlation coefficient (ICC) 0.69-0.88) and moderate-to-good test-retest reliability (ICC 0.56-0.75). Bland-Altman analysis indicated good repeatability. Normative percentiles for letter-based CSF were established.
CONCLUSIONS: CSF outcomes depend on stimulus type when measured on the same investigated LCD system. Grating stimuli are susceptible to ceiling effects, whereas letters demonstrate good reliability and provide clinically useful normative data, supporting their use in routine functional vision assessment.