Smith, A. E. | bioRxiv Science

bioRxiv2022

Resting EEG power spectra across middle to late life: Associations with age, cognition, APOE-ε4 carriage and cardiometabolic burden

We investigated how resting EEG measures are associated with risk factors for late-life cognitive impairment and dementia, including age, Apolipoprotein E {varepsilon}4 (APOE-{varepsilon}4) carriage and cardiometabolic burden. Resting EEG was recorded from 86 adults (50-80 years of age). Participants additionally completed the Addenbrookes Cognitive Examination (ACE) III and had blood drawn to assess APOE-{varepsilon}4 carriage status and cardiometabolic burden. EEG power spectra were decomposed into sources of periodic and aperiodic activity to derive measures of aperiodic component slope and alpha (7-14 Hz) and beta (15-30 Hz) peak power and peak frequency. Alpha and beta peak power measures were corrected for aperiodic activity. The aperiodic component slope was correlated with ACE-III scores but not age. Alpha peak frequency decreased with age. Individuals with higher cardiometabolic burden had lower alpha peak frequencies and lower beta peak power. APOE-{varepsilon}4 carriers had lower beta peak frequencies. Our findings suggest that the slope of the aperiodic component of resting EEG power spectra is more closely associated with measures of cognitive performance rather than chronological age in older adults.

neuroscience↗

bioRxiv2021

Age differences in aperiodic neural activity measured with resting EEG

Previous research using electroencephalography (EEG) and magnetoencephalography (MEG) has shown that neural oscillatory activity within the alpha band (8-12 Hz) becomes slower and lower in amplitude with advanced age. However, most studies have focused on quantifying age-related differences in periodic oscillatory activity with little consideration of the influence of aperiodic activity on these measures. The aim of this study was to investigate age differences in aperiodic activity inherent in the resting EEG signal. We assessed aperiodic activity in 85 healthy younger adults (mean age: 22.2 years, SD: 3.9, age range: 18-35, 37 male) and 92 healthy older adults (mean age: 66.1 years, SD: 8.2, age range 50-86, 53 male) by fitting the 1/f-like background activity evident in EEG power spectra using the fitting oscillations & one over f (FOOOF) toolbox. Across the scalp, the aperiodic exponent and offset were smaller in older compared to younger participants, reflecting a flatter 1/f-like slope and a downward broadband shift in the power spectra with age. Before correcting for aperiodic activity, older adults showed slower peak alpha frequency and reduced peak alpha power relative to younger adults. After correcting for aperiodic activity, peak alpha frequency remained slower in older adults; however, peak alpha power no longer differed statistically between age groups. The large sample size utilized in this study, as well as the depth of analysis, provides further evidence that the aperiodic component of the resting EEG signal is altered with aging and should be considered when investigating neural oscillatory activity.

neuroscience↗