Frontiers in Neurology · Oct 31, 2026
A neuropsychological battery for use in sleep disorders: Establishment of normative data and potential clinical thresholds
By Lucie Barateau, Robert H Pietrzak, Yves A Dauvilliers et al.
Objective: Cognitive symptoms, particularly in attention, are a common but under-recognized clinical feature of sleep disorders. This observational study aimed to establish normative benchmarks for cognitive tests assessing neuropsychological constructs of attention, memory, and executive function and derive impairment thresholds for clinical and research applications. Methods: A total of 127 cognitively unimpaired adults (age 18−75 years, 54% female) without excessive daytime sleepiness, marked self-reported cognitive difficulties, mood or neurological disorders, or untreated medical conditions affecting cognition completed the Psychomotor Vigilance Test, Sustained Attention to Response Test, Continuous Paired Associate Learning test, One-Back Test, and the International Digit Symbol Substitution Test-Symbols. Associations with demographic characteristics were examined using negative binomial models for count outcomes and Gaussian linear models for continuous outcomes. Primary abnormality thresholds were defined from age-adjusted standardized residuals using empirical 90th, 95th, and 98th percentile criteria in the direction of worse performance. Battery-level false-positive rates were calculated directly from the observed joint distribution of abnormal scores. Results: In the healthy adults, increasing age was associated with poorer performance on the Continuous Paired Associate Learning test and International Digit Symbol Substitution Test-Symbols. No significant demographic associations were observed for the Psychomotor Vigilance Test, Sustained Attention to Response Test, or One-Back Test. Based on age-adjusted empirical percentile thresholds, requiring at least three abnormal scores at the 90th percentile criterion, at least two at the 95th percentile criterion, or at least two at the 98th percentile criterion yielded observed specificities of 96.9%, 94.5%, and 100.0%, respectively. Conclusions: These normative data provide age-adjusted, distribution-free thresholds for interpreting performance on a brief neuropsychological battery. Empirically calibrated multivariate rules may help limit false-positive classifications when the battery is applied to adults with sleep disorders, although independent validation in clinical samples is needed.