Interactions between age and sex in multiscale entropy and spectral power changes across the lifespan

Multiscale entropy (MSE) changes in relation to age, whereby ageing is associated with an increasing bias towards fine scale entropy. This change is thought to represent a shift towards localized information processing in the brain as we age. However, this relationship has not been tested in large sample sizes alongside other demographic factors and cognitive behaviours. This study aimed to replicate previously reported effects of ageing on MSE in an open access database (Cambridge Centre for Ageing and Neuroscience, N = 587) and expand the findings to include an investigation of the effects of sex and a variety of cognitive behaviours. MSE curves and power spectrum densities (PSD) were calculated for each region of interest from the magnetoencephalography data. Multivariate partial least squares analyses were used to assess the relationship between MSE or PSD and five behavioural/demographic factors including age, sex, fluid intelligence, visual short-term memory, and a generalized measure of cognitive function. Age was associated with increased fine scale and decreased coarse scale entropy, as well as complementary spectral changes, including slowing of peak alpha rhythms, increased beta-band activity, and reduced gamma-band activity, which replicates prior MSE and PSD findings. In both domains, these age-related patterns differentiated based on sex with advancing age. Importantly, the unique effects of sex diverged between MSE and PSD. This result indicates that entropy-based measures can isolate aspects of temporal organization that are not clearly summarized by spectral structure alone.