As we age, the anatomy and physiology of our brain decline, and so do our cognitive abilities. As our brain shrivels and its functions dwindle, cognitive processes slow down and fail. Yet, brain and cognitive systems do not endure the assaults of aging passively; they respond plastically to these attacks by reorganizing their functions. Functional neuroimaging techniques, such as positron emission tomography (PET) and functional MRI (fMRI), provide ideal methods to investigate patterns of neurocognitive decline and compensation in older adults. During cognitive performance, older adults (OAs) often fail to activate some of the brain regions recruited by young adults (YAs), and instead recruit other brain regions. These age-related changes in brain activity can be linked directly to the effects of aging on behavioral measures, providing a bridge between cerebral aging and cognitive aging. For this reason, the field of functional neuroimaging of cognitive aging has grown very fast since the mid-1990s. The goal of this chapter is to provide a concise introduction to this new and exiting field. As an introduction, we briefly describe two consistent patterns of age-related changes in brain activity. Then we review PET and fMRI studies of aging in various cognitive domains. Finally, we discuss some current issues and future directions.

Introduction

Functional neuroimaging studies have revealed at least two consistent patterns of age-related changes in brain activity during cognitive performance. First, several studies, particularly in the visual perception domain, have found an age-related decrease in occipital activity coupled with an age-related increase in PFC activity. Grady et al. (1994) were the first to describe this occipital-decrease/frontal-increase (ODFI) pattern, and they suggested that OAs compensated for visual processing deficits (occipital decrease) by recruiting higher-order cognitive processes (PFC increase). In this study OAs and YAs were matched in accuracy but differed in reaction times (RTs), so the authors further suggested that additional recruitment of PFC functions allows OAs to maintain a good accuracy level at the expense of slower reaction times. Most subsequent studies that found OFDI endorsed Grady et al.'s compensatory account of age-related PFC increases.

A second consistent finding is a tendency of OAs to show more bilateral (less asymmetric) PFC activations than YAs. This pattern was conceptualized as a Hemispheric Asymmetry Reduction in Older Adults (HAROLD) model (Cabeza, 2002), and a sample of studies supporting the model are listed in table 12.1. The HAROLD pattern was originally described by Cabeza, Grady et al. (1997) and attributed to a compensatory mechanism. This compensation account is consistent with evidence that bilateral activity in OAs is positively correlated with successful cognitive performance (Reuter-Lorenz et al., 2000), and is found in high-performing rather than in low-performing OAs (Cabeza et al., 2002; Rosen et al., 2002). However, an alternative account is that a more widespread activation pattern reflects an age-related difficulty in engaging specialized neural mechanisms (e.g., Li & Lindenberger, 1999; Logan et al., 2002). This dedifferentiation account is consistent with an age-related increase in correlations across tasks (Lindenberger & Baltes, 1994). In general, available evidence tends to be more consistent with the compensation than with the dedifferentiation account (Daselaar & Cabeza, 2005), but further research is certainly required.