Uncorrected errors and correct saccades in the antisaccade task distinguish between early-stage Alzheimer’s disease dementia, amnestic mild cognitive impairment, and normal aging

Eraslan Boz H., Koçoğlu K., Akkoyun M., Tüfekci I. Y., Ekin M., Özçelik P., ...More

Aging, Neuropsychology, and Cognition, 2023 (SSCI) identifier identifier

  • Publication Type: Article / Article
  • Publication Date: 2023
  • Doi Number: 10.1080/13825585.2023.2198191
  • Journal Name: Aging, Neuropsychology, and Cognition
  • Journal Indexes: Social Sciences Citation Index (SSCI), Scopus, Academic Search Premier, Abstracts in Social Gerontology, AgeLine, CINAHL, EMBASE, Linguistics & Language Behavior Abstracts, MEDLINE, Psycinfo
  • Keywords: Alzheimer’s disease, amnestic mild cognitive impairment, Antisaccade, saccadic eye movements, uncorrected errors
  • Dokuz Eylül University Affiliated: Yes


Alzheimer’s disease (AD) dementia is a degenerative illness that is characterized by a gradual decline in cognitive abilities. Amnestic mild cognitive impairment (aMCI) is seen as a precursor to AD. The changes in antisaccade performance that can be seen in MCI may provide important clues in the early detection of AD. Therefore, the antisaccade deficits in AD and aMCI remain a research question. This study aimed to examine antisaccade responses and the relationship between antisaccade and cognitive function in AD, aMCI, and healthy controls (HC). This study included 30 patients with early-stage AD, 34 with aMCI, and 32 HC. Patients with AD showed higher rates of uncorrected error, anticipatory saccades and corrected errors, as well as decreased correct saccade rates, and shortened saccade latency compared to aMCI and HC in this study. Patients with aMCI exhibited increased rates of express saccades relative to HC. The antisaccade task and cognitive domains were found to be significantly related. Our study showed that the rate of correct saccades has the capacity to distinguish AD from HC with 87% sensitivity and 86% specificity (AUC = 0.93, p < 0.001). In addition, the rate of uncorrected errors was found to be capable of distinguishing AD from HC with 84% sensitivity and 83% specificity (AUC = 0.91, p < 0.001). This study presented promising findings that these parameters can be used clinically to differentiate AD and aMCI from healthy older individuals.