BackgroundSpatial memory and orientation deficits often precede cognitive impairment in incipient dementia, e.g., Alzheimer’s disease. Therefore, early diagnosis of spatial impairment may be crucial to the initiation of an adequate therapeutic intervention. Subjective tests, such as spatial anxiety and spatial discomfort questionnaires, and objective tests in the form of quantitative measures of orientation, are available. In these tests, vestibular hypofunction has often been neglected as a potential confounder. The major research question in this study was how self-assessed questionnaires correlate with the data from objective measures in participants with proven normal vestibular function.MethodsA heterogeneous group of 135 participants (72 females, 63 males, mean age 62.75 ± 14.46 years) from a tertiary center for vertigo and balance disorders consisting of two cohorts, with (n = 49) and without (n = 86) cognitive deficits in a screening test (MoCA), was examined (a) with a newly introduced inventory for subjective spatial discomfort (Extended Inventory for Spatial Orientation Discomfort, EISOD), (b) a well-established questionnaire for subjective spatial skills (Santa Barbara Sense of Direction Scale, SBSODS), and (c) the objective three-dimensional real-world pointing task (3D-RWPT) before and after horizontal body rotations. In all patients, acute central or peripheral vestibular deficits were ruled out by neuro-orthoptics, bithermal water calorics and video head impulse testing.ResultsSelf-assessed spatial orientation discomfort (EISOD) correlated with the amount of spatial impairment in the 3D-RWPT for both cohorts. The cognitively impaired patients showed significantly higher levels of spatial discomfort (i.e., lower scores; Welch’s t-test t-2.58, p < 0.01, Cohen’s d − 0.46), and higher angular deviations in the (cognitively demanding) transformation paradigm of the 3D-RWPT (t 2.37, p 0.02, Cohen’s d 0.44). They preferred retinotopic/egocentric spatial encoding strategies in the pointing task (Welch’s t-test t-2.61, p < 0.01, Cohen’s d − 0.47). In contrast, the self-report of spatial abilities (SBSODS) yielded no significant group differences (t − 1.66, p 0.10) and was not reliably associated with objective accuracy in the pointing task.ConclusionIn patients without vestibular deficits, subjective spatial discomfort (EISOD) correlated with the accuracy in an objective 3D-pointing task for both cohorts, and higher discomfort was associated with more severe cognitive impairment. EISOD-scores showed higher correlation indices than a self-report of spatial skills using the SBSODS. When investigating spatial abilities in patients with suspected cognitive impairment, it appears reasonable that both subjective spatial discomfort, subjective spatial abilities, and objective spatial measures should be combined. Future research in patients with vestibular dysfunction is needed to understand the role of vestibular deficits for the development of spatial orientation discomfort.