The role of working memory and attentional disengagement on inhibitory control: effects of aging and Alzheimer's disease

Abstract

Patients with Alzheimer's disease have an impairment of inhibitory control for reasons that are currently unclear. Using an eye-tracking task (the gap-overlap paradigm), we examined whether the uncorrected errors relate to the task of attentional disengagement in preparation for action. Alternatively, the difficulty in correcting for errors may be caused by the working memory representation of the task. A major aim of this study was to distinguish between the effects of healthy aging and neurodegenerative disease on the voluntary control of saccadic eye movements. Using the antisaccade task (AST) and pro-saccade task (PST) with the 'gap' and 'overlap' procedures, we obtained detailed eye-tracking measures in patients, with 18 patients with probable Alzheimer's disease, 25 patients with Parkinson's disease and 17 healthy young and 18 old participants. Uncorrected errors in the AST were selectively increased in Alzheimer's disease, but not in Parkinson's disease compared to the control groups. These uncorrected errors were strongly correlated with spatial working memory. There was an increase in the saccade reaction times to targets that were presented simultaneously with the fixation stimulus, compared to the removal of fixation. This 'gap' effect (i.e. overlap-gap) saccade reaction time was elevated in the older groups compared to young group, which yielded a strong effect of aging and no specific effect of neurodegenerative disease. Healthy aging, rather than neurodegenerative disease, accounted for the increase in the saccade reaction times to the target that are presented simultaneously with a fixation stimulus. These results suggest that the impairment of inhibitory control in the AST may provide a convenient and putative mark of working memory dysfunction in Alzheimer's disease.

Fig. 1

In the gap tasks, the fixation point is removed before the onset of the target, while in the overlap task the fixation point is visible when the target is presented. In the antisaccade task, the observer is required to saccade away from the target. The diagram illustrates the different types of errors that are generated in the gap and overlap paradigms. The green arrows indicate correctly directed saccade and fixations; red arrows show incorrect saccades and anticipations

Fig. 2

The mean gap effect (overlap–gap) and standard error bars for saccade reaction times in Alzheimer's disease (AD), Parkinson's disease (PD), old controls (OC) and young control (YC) groups

Fig. 3

a, b Mean uncorrected errors and standard error bars in gap (upper graph) and overlap (lower graph antisaccade tasks (AST). AD Alzheimer's disease; PD Parkinson's disease; OC old controls; YC young controls

Fig. 4

a, b Scatter plot of uncorrected antisaccade task errors (AST) against spatial working memory (spatial span reversed) and verbal working memory (digit span reversed scores). The r² values reveal a more robust relationship between spatial working memory and uncorrected AST errors (upper graph)