# Social cognition in Korsakoff's syndrome: A meta‐analysis

## TL;DR

People with Korsakoff's syndrome show significant deficits in social cognition compared to controls, especially in emotion perception and interpretation.

## Contribution

This is the first meta-analysis quantifying social cognition impairments in Korsakoff's syndrome across multiple domains.

## Key findings

- Individuals with Korsakoff's syndrome performed significantly worse in all social cognition domains compared to controls.
- The largest effect sizes were found in emotion perception and mentalizing tasks.
- Results for empathy and moral reasoning were inconclusive due to limited data.

## Abstract

Korsakoff's syndrome is an alcohol‐related neurocognitive disorder characterized by episodic memory impairments, apathy, confabulations and poor illness‐insight. This meta‐analysis aimed to estimate mean effect sizes of performance in social cognition in people with Korsakoff's syndrome (KS) compared with controls.

A systematic literature search was conducted in May 2024 to identify research articles that examined social cognition in patients with KS and control groups. Weighted effect sizes (Hedges' g) were calculated for the three levels of social cognition: emotion perception, social interpretation and socio‐cognitive integration. There was no restriction on setting. Instruments examining emotion perception (facial emotion recognition and prosody), interpretation (mentalizing, self‐awareness and empathy) and socio‐cognitive integration (moral reasoning and social knowledge) were used in the included studies.

Thirteen studies (n = 622; 292 KS, 330 controls) showed that individuals with KS performed statistically significantly worse across all domains of social cognition compared with controls. Large effect sizes were found in emotion perception [g = −1.14, 95% confidence interval (CI) = −1.46 to −0.81), P < 0.001, I2 = 58.5%, 8 studies, n = 372], with comparable effect sizes for facial emotion recognition and prosody. In social interpretation (g = −0.77, 95% CI = −1.34 to −0.21, P = 0.007, I2 = 96.6%, 4 studies, n = 188), a large effect was found for mentalizing (g = −1.05, 95% CI = −1.61 to −0.50, P < 0.001, I2 = 74.1%; 3 studies, n = 120). In socio‐cognitive integration (g = −0.74, 95% CI = −1.11 to −0.37, P < 0.001, I2 = 0%, 3 studies, n = 184), social knowledge showed a large effect size (g = −0.81, 95% CI = −1.24 to −0.38, P < 0.001, 1 study, n = 104). Results for empathy (g = −0.43, 95% CI = −1.05 to 0.20, P = 0.18, 1 study, n = 40), self‐awareness (g = −0.21, 95% CI = −0.47 to 0.05, P = 0.12, 1 study, n = 68) and moral reasoning (k = 2, g = −0.54, 95% CI = −1.28 to 0.19, P = 0.15, I2 = 0%; 2 studies, n = 80) were uncertain, with possible important differences in both directions.

This meta‐analysis shows that people with Korsakoff's syndrome perform statistically significantly worse than controls on socio‐cognitive measures, with the largest effect sizes in the perception and interpretation of social information.