Association between Contrast Sensitivity and Reading with Macular Pathology

SIGNIFICANCE: Comparison between the role of spatial and temporal contrast sensitivities in the association with reading may provide insight into how visual tasks (such as reading) are related to primary optical or neural (or both) effects. More insight into primary visual factors influencing reading is important for understanding reading problems.

PURPOSE: The aim of this study was to gain insight into the association between optical and neural components of contrast sensitivity (CS), operationalized as spatial CS (optical and neural) or temporal CS (solely neural), and reading speed in a clinical sample of participants with macular pathologies. The precision and agreement were also investigated.

METHODS: The Mars test and temporal CS implementation of the C-Quant device were used to measure spatial CS and temporal CS, respectively. Tests were performed with 47 participants: mean age, 77 years (range, 52 to 92 years). Associations were investigated with correlations and linear regression models. Precision was defined by coefficients of repeatability. The 95% limits of agreement between spatial CS and temporal CS values were assessed.

RESULTS: Reading speed correlated with both spatial CS (r = 0.35, P = .015) and temporal CS (r = 0.66, P < .001). After correction for visual acuity, central loss, and education level, the association between temporal CS and reading speed was not significant anymore. The coefficients of repeatability and reproducibility were 0.20 and 0.28 log unit (spatial CS) and 0.33 and 0.35 log unit (temporal CS), respectively. The values for temporal CS were 0.08 and 0.13 log unit higher than those for spatial CS.

CONCLUSIONS: For spatial CS and temporal CS, moderate to strong correlations were found, respectively, with reading speed in patients with maculopathies. The stronger association between temporal CS and reading speed is suggested to reflect a high sensitivity for neural integrity of temporal CS. The differences in coefficients of repeatability and reproducibility could be explained by the psychometrical differences between methods.