Background Facial emotion recognition (FER) is increasingly recognized as a functionally relevant domain in the Alzheimer’s disease (AD) spectrum, yet whether baseline FER is associated with subsequent clinical decline remains insufficiently established. Methods We studied 328 participants [subjective cognitive decline (SCD) n = 94, mild cognitive impairment (MCI) n = 142, AD dementia n = 92] from a longitudinal memory clinic cohort; 148 had follow-up assessments (mean 1.59 ± 0.67 years). FER was assessed using the Korean version of the Florida Affect Battery (K-FAB; five subtests, FER_mean). Linear mixed-effects models examined group-differential trajectories of the Mini-Mental State Examination (MMSE), Clinical Dementia Rating–Sum of Boxes (CDR-SOB), and FER_mean, adjusted for age, sex, education, and baseline values. Results FER_mean decreased stepwise across groups at baseline (SCD 17.3 ± 1.5, MCI 16.4 ± 2.1, AD 15.3 ± 2.0; p < 0.001). CDR-SOB showed more consistent group-differential longitudinal trajectories: MCI increased at +0.550 points/year ( p < 0.001), with SCD significantly slower ( β = −0.491, p = 0.008) and AD faster ( β = +0.371, p = 0.038). FER_mean showed a modest decline in MCI ( β = −0.169/yr., p = 0.078) and SCD showed significantly attenuated decline relative to MCI ( β = +0.376, p = 0.035). Lower baseline FER was associated with faster CDR-SOB worsening ( ρ = −0.221, p = 0.006) but not MMSE decline ( ρ = 0.093, p = 0.255). FER change rates showed a weak, non-significant correlation with MMSE change ( ρ = 0.004, p = 0.960) but a modest significant correlation with CDR-SOB change ( ρ = −0.195, p = 0.018), suggesting partially distinct longitudinal patterns. Conclusion Baseline FER was modestly associated with longitudinal worsening in clinical severity (CDR-SOB), but not cognitive screening performance (MMSE) across the pre-dementia to dementia spectrum. FER trajectories showed limited correlation with traditional measures. These findings indicate that FER may have a potential complementary role of social cognitive assessment in the AD continuum, pending further validation.
No takes yet. Share an insight, caveat, or question.
Shim et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: