The Stroop is a frequently used neuropsychological test, with poor performance typically interpreted as indicative of disinhibition and frontal lobe damage. This study tested those interpretations by examining relationships between Stroop performance, behavioral disinhibition, and frontal lobe atrophy.
Participants were 112 well-characterized patients with mild cognitive impairment or dementia, recruited through UCSF's Memory and Aging Center. Participants received comprehensive dementia evaluations including structural MRI, neuropsychological testing, and informant interviews. Freesurfer, a semi-automated parcellation program, was used to analyze 1.5T MRI scans. Behavioral disinhibition was measured using the Disinhibition scale of the Neuropsychiatric Inventory. The sample (n=112) mean age was 65.40 (SD=8.60) years, education was 16.64 (SD=2.54) years, and MMSE was 26.63 (SD=3.32). Hierarchical linear regressions were used for data analysis.
Controlling for age, MMSE, and Color Naming performance, Stroop performance was not significantly associated with behavioral disinhibition (β=0.01, ΔR2=0.01, p=0.29). Hierarchical regressions controlling for age, MMSE, Color Naming, intracranial volume, and temporal and parietal lobes, examined whether left hemisphere or right hemisphere regions predict Interference speed. Bilaterally, parietal lobes were the brain region in which atrophy best predicted poorer Stroop (left: β=0.0004, ΔR2=0.02, p=0.002; right: β=0.0004, ΔR2=0.02, p=0.002). Of frontal regions, only dorsolateral prefrontal cortex atrophy predicted poorer Stroop (β=0.001, ΔR2=0.01, p=0.03); left and right anterior cingulate cortex (ACC) atrophy predicted better Stroop (left: β=−0.003, ΔR2=0.01, p=0.02; right: β=−0.004, ΔR2=0.01, p=0.02).
These findings suggest Stroop performance is a poor measure of behavioral disinhibition and frontal lobe atrophy even among a relatively high-risk population.