P2-418: association between heart rate variability and bold fmri signal variability in younger and older adults: comparison between resting state and task

Greater signal variability in the Blood Oxygen-Level Dependent (BOLD) signal from fMRI has been associated with optimal brain function, including flexibility/adaptability. Heart rate variability (HRV) is associated with important aspects of self-regulation, including physiological, cognitive, and emotion regulation. High-frequency (HF) HRV represents primarily parasympathetic influences and low-frequency (LF) HRV represents a mixture of sympathetic and parasympathetic influences. While both are known to decrease with age, it remains unclear whether and how they are associated with each other. Across two studies, we addressed the following questions: 1) Is HRV associated with BOLD variability? 2) Are HF-HRV and LF-HRV differently associated with BOLD variability depending on the task type? 3) Does this association change with age? In Study 1, we measured the baseline HRV from 19 older adults and 16 younger adults. We then administered the resting-state and task-based (emotion encoding) fMRI scans to measure BOLD signal variability. In Study 2, 19 older adults and 23 younger adults completed two task-based fMRI scans (fear conditioning, attention task) while heart rates were recorded. For the brain signal variability, we measured the standard deviation of the BOLD signal (SDBOLD). We examined SDBOLD from whole brain, several regions involved in emotional regulation process and 18 intrinsic network masks. For heart rate variability, we measured the HF-power and LF-power from frequency analysis. In Study 1, entire sample showed that SDBOLD in the resting state significantly positively correlated with HF-HRV and SDBOLD for the task activity the significantly positively correlated with LF-HRV. After controlling for age, the correlations remained. But only older group showed the significant correlations. In Study 2, entire sample showed that SDBOLD for the task activity significantly positively correlated with LF-HRV. After controlling for age, the correlations remained. But only younger group showed the significant correlations. The present results suggest that resting-state and task-based BOLD variabilities are differently related to parasympathetic versus sympathetic system influences and that the association between HRV and BOLD signal variability changes with aging.