Inhibitory Control of Saccadic Eye Movements and Cognitive Impairment in Mild Cognitive Impairment

Abstract

Background: Mild cognitive impairment (MCI) may occur due to several forms of neurodegenerative diseases and non-degenerative conditions and is associated with cognitive impairment that does not affect everyday activities. For a timely diagnosis of MCI to prevent progression to dementia, a screening tool of fast, low-cost and easy access is needed. Recent research on eye movement hints it a potential application for the MCI screening. However, the precise extent of cognitive function decline and eye-movement control alterations in patients with MCI is still unclear.

Objective: This study examined executive control deficits and saccade behavioral changes in patients with MCI using comprehensive neuropsychological assessment and interleaved saccade paradigms.

Methods: Patients with MCI (n = 79) and age-matched cognitively healthy controls (HC) (n = 170) completed four saccadic eye-movement paradigms: prosaccade (PS)/antisaccade (AS), Go/No-go, and a battery of neuropsychological tests.

Results: The findings revealed significantly longer latency in patients with MCI than in HC during the PS task. Additionally, patients with MCI had a lower proportion of correct responses and a marked increase in inhibition errors for both PS/AS and Go/No-go tasks. Furthermore, when patients with MCI made errors, they failed to self-correct many of these inhibition errors. In addition to the increase in inhibition errors and uncorrected inhibition errors, patients with MCI demonstrated a trend toward increased correction latencies. We also showed a relationship between neuropsychological scores and correct and error saccade responses.

Conclusion: Our results demonstrate that, similar to patients with Alzheimer's dementia (AD), patients with MCI generate a high proportion of erroneous saccades toward the prepotent target and fail to self-correct many of these errors, which is consistent with an impairment of inhibitory control and error monitoring.

Significance: The interleaved PS/AS and Go/No-go paradigms are sensitive and objective at detecting subtle cognitive deficits and saccade changes in MCI, indicating that these saccadic eye movement paradigms have clinical potential as a screening tool for MCI.

Keywords: Go/No-go; frontal/executive function; inhibitory control; mild cognitive impairment; prosaccade/antisaccade; self-monitoring.

FIGURE 1

Cue and target stimuli for PS/Go, AS, and No-go trials.

FIGURE 2

PS/AS and Go/No-go paradigms.

FIGURE 3

Oculomotor responses in controls and patients with MCI. (A) Proportion for prosaccade trials. (B) Proportion for antisaccade trials. (C) proportion for go trials. (D) Proportion for no-go trials. (E) Latency for PS, AS, Go trials, and proportion and fixation duration for no-go trials. **Significance level p < 0.001; *significance level p < 0.05; n.s., not significant.

FIGURE 4

Correlation of PS/AS responses and neuropsychological tests for (A) PS responses by HC. (B) PS responses by MCI patients. (C) AS responses by HC. (D) AS responses by MCI patients. Data were presented with the absolute values of Pearson correlation coefficients; exact values and p-values are presented in Supplementary Table 2 (HC) and Supplementary Table 3 (MCI). SC, self-corrected errors; SCL, self-corrected latency; SCT, self-corrected time; UCI, uncorrected-inhibition errors; CI, corrected inhibition errors; CIL, corrected inhibition latency; CIT, corrected inhibition time. Weak correlations, near 0, are in white while those nearing 1 are in red, portraying strong correlations. Pearson’s correlations are in absolute values.

FIGURE 5

Correlation of Go/No-Go responses and neuropsychological tests for (A) Go responses by HC. (B) Go responses by MCI patients. (C) No-Go responses by HC. (D) No-Go responses by MCI patients. Data were presented with the absolute values of Pearson correlation coefficients; exact values and p values are presented in Supplementary Table 2 (HC) and Supplementary Table 3 (MCI). FD, fixation duration, SC, self-corrected errors, SCL, self-corrected latency, SCT, self-corrected time, UCI, uncorrected-inhibition errors, CI, corrected inhibition errors, CIL, corrected inhibition latency, and CIT, corrected inhibition time. Weak correlations, near 0, are in white while those nearing 1 are in red, portraying strong correlations. Pearson’s correlations are in absolute values.