The effects of aging and cognitive performance on patterns of neural activity measured by functional magnetic resonance imaging

Abstract

Previous functional magnetic resonance imaging (fMRI) research has shown that older adults activated bilateral frontal regions during tasks in which young adults had unilateral frontal activation (Cabeza, 2001). It has been suggested that older adults recruit bilateral frontal regions to compensate for declining brain function in other regions (Cabeza, Anderson, Locantore, and McIntosh, 2002). The primary aim of the current study was to determine how bilateral activation patterns observed in the frontal lobe during encoding and recognition were related to both cognitive performance of older adults and to function in other brain regions. Thirty-five older adults and 9 young adults completed an encoding and recognition task during fMRI scanning. During the encoding scans participants determined whether presented words were "natural" or "man made" objects. During the recognition scans, participants made "old/new" judgments for each word presented. Four sets of bilateral regions of interest (ROI) were defined from an overlap image of all participants' fMRI data: (1) right and left frontal cortex, (2) right and left medial temporal lobe, (3) right and left parietal lobe, and (4) right and left lateral temporal lobe. On a separate day participants completed a neuropsychological testing session that included a series of tests that had been previously used to characterize older adults in two cognitive domains, frontal (FL) and medial temporal lobe (MTL) function (Glisky, Polster & Routhieaux, 1995; Glisky, Rubin & Davidson, 2001). Consistent with the previous research, older adults showed greater bilateral fMRI activation in the frontal lobes during encoding than young adults. However, bilateral activation in the frontal lobes during encoding was associated with two different activation patterns: (1) when MTL activation was present, bilateral frontal activation was observed in older adults with high FL factor scores; (2) when the MTL was not active, bilateral frontal activation was found in older adults with low MTL factor scores. Older adults with high FL factor scores but who did not activate MTL had left lateralized frontal activation. Importantly, older adults with and without MTL activation did not differ in recognition performance scores, or factor scores.