Inhibitory Control Changes in Healthy Aging, Alzheimer’s Disease and Amnestic Mild Cognitive Impairment: An FMRI Study of the Stroop Task
Alzheimer's Disease , Amnestic Mild Cognitive Impairment , Healthy Aging , Stroop Task , Inhibitory Control
Alzheimer’s disease (AD) is a neurodegenerative disorder characterized by impairments in cognition coupled with a predictive pattern of neural atrophy beyond changes associated with healthy aging. Divergence from normative aging occurs years before clinical diagnosis of AD, a prodromal period classified as amnestic mild cognitive impairment (aMCI). Inhibitory control has been investigated across healthy aging, AD and aMCI through utilization of the Stroop task. Inhibition is required to suppress the automatic reading response in order to identify the ink color of word stimuli with matching (congruent) or non-matching (incongruent) ink color. Whether differences exist behaviorally, or if compensatory brain mechanisms to maintain task performance occur across these groups remains unclear. To address this knowledge gap, we utilized functional magnetic resonance imaging (fMRI) to evaluate the relationship between behavioral measures, neural correlates, and functional connectivity during performance of the Stroop task in young, middle-aged, and older adults, as well as patients with AD and aMCI. We found evidence of over-recruitment in middle-aged adults, which became more discernable after age 65 with both over-recruitment and loss of specificity to stimuli. Despite changes detected using fMRI, behavioral performance was similar across groups, with generalized slowing in older adults. Patients with aMCI performed similarly to controls, however showed reduced BOLD response within the parietal ROIs and over-recruitment of frontal ROIs when faced with incongruent trials. Patients with AD did not elicit different activity from controls in any ROI but did produce significantly more incongruent errors. Patients with AD had reduced functional connectivity in frontal brain areas, which correlated with default mode network (DMN) seeds. Functional connectivity was similar between aMCI and controls, suggesting that functional connectivity of Stroop and DMN ROIs remain spared early in the disease. Our results suggest that reaction times alone may not be sensitive enough to show changes in aging or aMCI, but incongruent errors may be particularly useful at discriminating AD. Our studies support the literature indicating compensation, as we found evidence of neural plasticity in both healthy and pathological aging. Methodological factors including Stroop design, fMRI analysis, and participant variability may contribute to inconsistencies within the literature.