Cardiorespiratory fitness is associated with fMRI signal in right cerebellum lobule VIIa Crus I and II during spatial navigation in older adult women

Abstract

Spatial navigation is a cognitive skill critical for accomplishing daily goal-directed behavior in a complex environment; however, older adults exhibit marked decline in navigation performance with age. Neuroprotective interventions that enhance the functional integrity of navigation-linked brain regions, such as those in the medial temporal lobe memory system, may preserve spatial navigation performance in older adults. Importantly, a well-established body of literature suggests that cardiorespiratory fitness has measurable effects on neurobiological integrity in the medial temporal lobes, as well as in other brain areas implicated in spatial navigation, such as the precuneus and cerebellum. However, whether cardiorespiratory fitness modulates brain activity in these regions during navigation in older adults remains unknown. Thus, the primary objective of the current study was to examine cardiorespiratory fitness as a modulator of fMRI activity in navigation-linked brain regions in cognitively healthy older adults. To accomplish this objective, cognitively intact participants (N = 22, aged 60-80 years) underwent cardiorespiratory fitness testing to estimate maximal oxygen uptake ( $\stackrel{·}{\text{V}}$ O_2max) and underwent whole-brain high-resolution fMRI while performing a virtual reality navigation task. Our older adult sample demonstrated significant fMRI signal in the right and left retrosplenial cortex, right precuneus, right and left inferior parietal cortex, right and left cerebellum lobule VIIa Crus I and II, right fusiform gyrus, right parahippocampal cortex, right lingual gyrus, and right hippocampus during encoding of a virtual environment. Most importantly, in women but not men (N = 16), cardiorespiratory fitness was positively associated with fMRI activity in the right cerebellum lobule VIIa Crus I and II, but not other navigation-linked brain areas. These findings suggest that the influence of cardiorespiratory fitness on brain function extends beyond the hippocampus, as observed in other work, to the cerebellum lobule VIIa Crus I and II, a component of the cerebellum that has recently been linked to cognition and more specifically, spatial processing.

Keywords: aging; cardiorespiratory fitness; cerebellum; fMRI; navigation.

FIGURE 1

Overview of study procedures. (A) Participants came in for three study visits: an eligibility visit, a fitness visit, and an MRI visit. (B) Virtual reality navigation task: encoding phase (inside the MRI scanner). During the “encoding” condition of the task, participants were instructed to navigate through a series of rooms and interconnected hallways with the objective of learning the location of six objects placed within the environment. During the “control” condition of the task, participants were instructed to move through a hallway as quickly as possible while also traveling over a series of floor markers. The task alternated between the “encoding” and “control” condition every 60 s for a total of 10 min. Participants completed this phase of the task using a four-button, diamond-configuration response device to move in the virtual space. (C) Virtual reality navigation task: retrieval phase (outside the scanner). Participants began at one object in the virtual environment and were instructed to navigate to another object using the shortest possible route. Upon correctly locating the specified object, the participant was instructed to find another object. This continued throughout the duration of the 60-s recall block. Each participant completed five 60-s recall blocks consecutively.

FIGURE 2

Distribution of cardiorespiratory fitness, as measured by estimated $\stackrel{·}{\text{V}}$O_2max, separated by sex. In women (N = 16), the mean ± SD estimated $\stackrel{·}{\text{V}}$O_2max was 28.95 ± 6.50 mL/kg/min, whereas in men (N = 6), the mean ± SD estimated $\stackrel{·}{\text{V}}$O_2max was 35.53 ± 4.72 mL/kg/min.

FIGURE 3

Clusters that demonstrated statistically significant activation (p < 0.001) during the navigation task in encoding > control conditions (orange) and control > encoding conditions (blue) as measured using a standard univariate analysis in AFNI (3dttest++ with ETAC). All clusters shown here survived correction for multiple comparisons and are demonstrated on the age-specific MRI brain template warped to MNI space. Note that all four panels (A–D) show a significant encoding > control cluster in three planes, listed left to right: axial (left = right hemisphere), sagittal (left = anterior), and coronal (left = right hemisphere). (A) Significant cluster comprising the bilateral retrosplenial cortex and right precuneus. (B) Two significant clusters comprising bilateral inferior parietal cortex. (C) Two significant clusters comprising the bilateral cerebellum (lobule VIIa Crus I and II). (D) Significant cluster comprising the medial temporal areas, including right fusiform gyrus, right parahippocampal gyrus (and more specifically, parahippocampal cortex), right lingual gyrus, and right hippocampus.

FIGURE 4

Cardiorespiratory fitness is associated with right cerebellum lobule VIIa Crus I and II beta coefficient in women. (A) Estimated $\stackrel{·}{\text{V}}$O_2max is significantly associated with the beta coefficient of the cluster in the right cerebellum lobule VIIa Crus I and II (estimated $\stackrel{·}{\text{V}}$O_2max—β = 0.023, SE = 0.006, t[13] = 3.881, p = 0.002**, p_adjust for ROIs (7) = 0.013*, delta-R² = 0.532). (B) Significant cluster comprised of the right cerebellum lobule VIIa Crus I and II shown in three planes, listed in order of top left, top right, and bottom: axial (left = right hemisphere), coronal (left = right hemisphere), and sagittal (left = anterior).