The present study evaluated the performance of the digital clock drawing test (dCDT) signature variable in assessing cognitive function in patients with multiple sclerosis (MS) and neuromyelitis optica spectrum disorder (NMOSD) compared to healthy controls (HC). Thirty-three participants, including 18 patients with MS or NMOSD and 15 HCs, underwent dCDT and Montreal Cognitive Assessment (MOCA) testing. The cognitive functions of the patients were further evaluated using a neuropsychological test battery. Global features based on time, space, and motion in the dCDT process were collected. ANOVA showed that consistent with MOCA scores, the three groups (HC, cognitively impaired patients, and cognitively normal patients) exhibited differences in CDT scores, clock digit placement, line length, and pen trajectory velocity characteristics (all p < 0.004). The follow-up work used correlation analysis and stepwise regression to assess the correlation between dCDT parameters and EDSS scores as well as neuropsychological tests. Notably, the vibration amplitude of pen movement during drawing ly correlated with the EDSS score (p < 0.001), indicating that it reflects clinical status. Other variables, such as the characteristic variable that characterizes the degree of spectral dispersion in pen movement trajectories, trajectory length, test duration, and average stress, significantly correlate with neuropsychological test results involving memory, language, and information processing speed at different levels (all p < 0.04). Our findings suggest that dCDT’s complex features link to cognitive abilities and physiological impairments in MS/NMOSD patients. Kinematic, temporal, and visual space-based global and process characteristics serve as valuable neurocognitive biomarkers. This offers the potential for dCDT to serve as a cognitive assessment tool for MS/NMOSD patients in the future.