Background: Event-related potentials (ERPs) and time-frequency analysis investigate cognition-related brain activity. The Symbol-Digit Modalities Test (SDMT) assesses cognitive processing in people with Multiple Sclerosis (pwMS). EEG may reveal neurophysiological dynamics during SDMT, aiding in identifying electrophysiological biomarkers of cognitive impairment. Methods: We developed an EEG-SDMT paradigm, presenting a symbol-number legend and 180 symbol-digit pairs. Reaction times (RTs) and 32-channel EEG data were recorded from 75 pwMS (50 F; 53.36 ± 10.03 years; EDSS 5.95 ± 1.01) and 22 healthy controls (HC) (15 F; 51.36 ± 13.88 years). Results: PwMS showed longer SDMT-ERP peak latency (612.87 ± 111.21 ms) vs HC (520.44 ± 79.99 ms) (p < 0.001). The cognitive processing component was delayed in pwMS (460.99 ± 117.23 ms) vs HC (376.39 ± 73.80 ms) (p < 0.001). Both latencies correlated with EEG-SDMT RTs (r = 0.278, p < 0.01; r = 0.214, p < 0.05). PwMS exhibited reduced alpha (200-300 ms) and theta (300−600 ms) activity (p < 0.01) in centro-parietal regions. EEG-SDMT RTs inversely correlated with SDMT scores (r = −0.718, p < 0.01). Conclusions: Our EEG-SDMT paradigm reveals neurophysiological correlates of SDMT performance, significantly impaired in pwMS, mainly due to cognitive rather than visual processing differences.