Baseline Oculomotor Parameters Are Prospectively Associated with Cognitive Processing Speed at 6-Year Follow-Up in Multiple Sclerosis: An Exploratory Cohort Study

Background/Objectives: Multiple sclerosis (MS) commonly impairs information processing speed, which is not well captured by conventional disability metrics. Oculomotor paradigms engage neural circuits frequently affected in MS and may provide objective cognitive correlates. We investigated whether baseline oculomotor parameters are prospectively associated with processing speed at ~6-year follow-up. Methods: Forty-four patients with MS underwent a standardized oculomotor battery (visually guided saccades, antisaccades, and sinusoidal smooth pursuit). The Symbol Digit Modalities Test (SDMT) was reassessed after a mean of 5.7 years. Age-adjusted Spearman partial correlations were computed between 30 baseline oculomotor parameters and follow-up SDMT, applying false discovery rate (FDR) correction. Sensitivity analyses included Cook’s distance. Results: Three parameters survived FDR correction: reflexive saccade duration (i.e., erroneous fixation duration—the dwell time at the erroneous target location before corrective antisaccade initiation; ρ = −0.59, q = 0.0011), catch-up saccades (ρ = −0.52, q = 0.0055), and reflexive saccade latency (ρ = −0.44, q = 0.033). Results remained stable after excluding influential observations. In multivariable analysis (adjusted R2 = 0.60), reflexive saccade duration (p < 0.001) and catch-up saccades (p = 0.019) were independently associated with lower SDMT. Conclusions: Baseline antisaccade reflexive saccade duration and smooth-pursuit catch-up saccades were prospectively associated with worse cognitive processing speed at long-term follow-up, suggesting eye-tracking-derived metrics as candidate objective correlates warranting prospective validation in MS.