Social cognition differs between NMOSD and multiple sclerosis, study finds

Social cognition, the capacity to perceive and interpret social information, is often impaired in neurological disorders such as multiple sclerosis, but its status in neuromyelitis optica spectrum disorder, a distinct autoimmune demyelinating disease, has remained unclear. A new cross-sectional study published in Discover Mental Health compared the two conditions directly, alongside healthy controls, using standardized measures of affective theory of mind, empathy, mood, fatigue, psychological distress and intelligence.

The research team recruited 107 participants: 32 people with neuromyelitis optica spectrum disorder, 38 with relapsing-remitting multiple sclerosis and 37 healthy controls. Each participant completed the Reading the Mind in the Eyes Test, the Empathy Quotient, the Beck Depression Inventory, the Modified Fatigue Impact Scale, the Kessler Psychological Distress Scale and the Wechsler Adult Intelligence Scale. Group comparisons were analyzed using analysis of variance and analysis of covariance, with covariate adjustment applied to account for differences between the groups.

The headline result concerns the Reading the Mind in the Eyes Test. Participants with neuromyelitis optica spectrum disorder scored lower than healthy controls, with mean scores of 16.81 plus or minus 4.29 versus 19.35 plus or minus 3.49, a difference that reached statistical significance at p equal to 0.02 in the unadjusted comparison. However, once the researchers adjusted for covariates, this difference was no longer significant, with an adjusted p value of 0.45. In other words, the apparent deficit in reading emotional states from the eyes appears to be explained by other factors that differed between the groups rather than by the disease itself.

Empathy told a different story. On the Empathy Quotient, the three groups did not differ significantly in the initial unadjusted analysis: the neuromyelitis optica group averaged 41.97 plus or minus 10.769, the relapsing-remitting multiple sclerosis group averaged 38.29 plus or minus 9.03, and controls averaged 42.05 plus or minus 9.39, with p equal to 0.17. Yet the adjusted analyses revealed a pattern the raw scores had obscured: empathy scores were lower in the relapsing-remitting multiple sclerosis group than in the neuromyelitis optica group, at p equal to 0.03. This reversal underscores why covariate adjustment matters in studies of clinical populations, where variables such as mood, fatigue, distress and intellectual function can differ between diagnostic groups and can mask or mimic disease-specific effects.

Intelligence quotient showed the clearest unadjusted separation among the three groups. Scores on the Wechsler Adult Intelligence Scale were significantly lower in the neuromyelitis optica group compared with both the multiple sclerosis group and controls: 76.68 plus or minus 43.17, 109.71 plus or minus 29.23 and 120.14 plus or minus 31.27 respectively, with p less than 0.0001. The very large standard deviation in the neuromyelitis optica group suggests considerable variability in performance among those participants.

Taken together, the findings suggest that social cognition impairments may differ between the two demyelinating disorders rather than following a single shared pattern. The researchers conclude that relapsing-remitting multiple sclerosis is associated with lower adjusted empathy quotient scores than neuromyelitis optica spectrum disorder, whereas the initial difference in affective theory-of-mind performance between the neuromyelitis optica group and healthy controls was not retained after covariate adjustment. This dissociation, between a measure that lost significance after adjustment and another that gained it, is the study’s central contribution.

The design carries important limitations. As a cross-sectional investigation, it captures a single time point and cannot establish whether the observed differences emerge, persist or change over the course of either disease. The sample sizes are modest, and the covariates used for adjustment may not capture every factor influencing social cognitive performance. The authors note that the findings warrant longitudinal and neuroimaging studies to clarify the mechanisms behind the differing social cognition profiles.

For clinicians and researchers, the practical implication is that social cognitive assessment in demyelinating disease may need to be tailored to the specific diagnosis. A deficit apparent on one measure in one condition may vanish under statistical adjustment, while a difference invisible in raw group means may emerge once confounding variables are controlled. The study, published as open access research on 02 November 2026, was conducted by a team including Iman Adibi, Zeynab Jalilian, Zeynab Zamani, Ali Rahmati, Omid Mirmosayyeb, Yasaman Shojaei, Mohammadamin Mazidi, Molod Mojodi, Fereshteh Ashtari, Vahid Shaygannejad and Armin Adibi. The article was shared early to provide faster access to peer-reviewed, accepted research, carries a permanent DOI and is citable, with the final version of record to follow.

Subject of Research: Social cognition, empathy and theory of mind in neuromyelitis optica spectrum disorder compared with relapsing-remitting multiple sclerosis and healthy controls

Article Title: Social cognition in neuromyelitis optica spectrum disorder compared with relapsing-remitting multiple sclerosis and healthy controls

Article References: Adibi, I., Jalilian, Z., Zamani, Z., Rahmati, A., Mirmosayyeb, O., Shojaei, Y., Mazidi, M., Mojodi, M., Ashtari, F., Shaygannejad, V., & Adibi, A. (2026). Social cognition in neuromyelitis optica spectrum disorder compared with relapsing-remitting multiple sclerosis and healthy controls. Discover Mental Health. https://doi.org/10.1007/s44192-026-00649-4

Image Credits: AI Generated

DOI: 10.1007/s44192-026-00649-4

Keywords: social cognition, neuromyelitis optica spectrum disorder, relapsing-remitting multiple sclerosis, theory of mind, empathy, Reading the Mind in the Eyes Test, Empathy Quotient, cross-sectional study, autoimmune demyelinating disease, covariate adjustment, intelligence quotient, healthy controls