Aperiodic EEG activity is a domain-specific marker of attentional aging in neurotypical adults

Aperiodic neural activity has been reliably shown to be reduced among older adults, and these reductions are associated with age-related cognitive decline. These results suggest that aperiodic activity, as measured by the spectral exponent, might be a viable electrophysiological biomarker for cognitive aging. However, this has not been tested longitudinally, within participants, during normal aging. Here, we used a longitudinal cohort of either sex to test whether changes in aperiodic activity between two sessions, approximately 5 years apart, track behavioral change across two domains: sustained attention (Psychomotor Vigilance Test) and executive function (modified Simon task). We found that aperiodic activity declined within individuals after the interval, independent of age at baseline. This within-person change tracked concurrent changes in reaction times during sustained attention but was not associated with changes in any Simon task measure. Within a resampling framework, the aperiodic-PVT association was stable across cohort resamples and independently fitted partitions, though it was not confirmed in a small, frozen held-out set. These findings suggest the longitudinal decline in aperiodic activity tracks attentional processing speed rather than executive function, supporting its potential as a domain-specific biomarker of attentional aging in neurotypical populations.Significance Statement Aperiodic (non-oscillatory) neural activity, indexed by the aperiodic exponent, is known to decline in older adults, but whether this happens within individuals over just a few years, and in step with cognitive change, remains unclear. Here we leveraged scalp EEG and cognitive-behavioral data from 276 neurotypical adults, tested twice approximately five years apart, and found that the aperiodic exponent declines within individuals over this interval, independent of age at baseline. This decline tracked concurrent slowing on a sustained attention task, but not change on an executive function task, and only in EEG recorded before, not after, cognitive testing. Based on this dissociation, we propose that aperiodic exponent decline indexes attentional processing speed rather than general cognitive decline in neurotypical aging.