Periodic and aperiodic neural activity displays age-dependent changes across early-to-middle childhood

Abstract

The neurodevelopmental period spanning early-to-middle childhood represents a time of significant growth and reorganisation throughout the cortex. Such changes are critical for the emergence and maturation of a range of social and cognitive processes. Here, we utilised both eyes open and eyes closed resting-state electroencephalography (EEG) to examine maturational changes in both oscillatory (i.e., periodic) and non-oscillatory (aperiodic, '1/f-like') activity in a large cohort of participants ranging from 4-to-12 years of age (N = 139, average age=9.41 years, SD=1.95). The EEG signal was parameterised into aperiodic and periodic components, and linear regression models were used to evaluate if chronological age could predict aperiodic exponent and offset, as well as well as peak frequency and power within the alpha and beta ranges. Exponent and offset were found to both decrease with age, while aperiodic-adjusted alpha peak frequency increased with age; however, there was no association between age and peak frequency for the beta band. Age was also unrelated to aperiodic-adjusted spectral power within either the alpha or beta bands, despite both frequency ranges being correlated with the aperiodic signal. Overall, these results highlight the capacity for both periodic and aperiodic features of the EEG to elucidate age-related functional changes within the developing brain.

Keywords: Aperiodic activity; EEG; Neurodevelopment; Neurophysiology; Oscillations; Spectral power.

Fig. 1

A) Example FOOOF model fit from a single subject showing the aperiodic exponent and offset (marked in red) across the analysed frequency range (1–40 Hz). The centre frequency, power, and bandwidth are highlighted (arrows) for the oscillatory peak present within the alpha range. B) Graphical illustration demonstrating shifts in the aperiodic exponent and offset. C) Aperiodic exponent and offset values for the eyes open (EO) and eyes closed (EC) recordings. Values are the average across all EEG electrodes. D) R-squared and error values for the model fit for the eyes open and eyes closed recordings (average across all electrodes). E) Topographic plots showing the spatial distribution of mean exponent and offset values across participants. Exponent values were highest near the midline, spanning frontal, central and posterior channels; while offset values were highest over posterior channels. F) Correlation between exponent and offset values (average over all electrodes). There was a strong association between both metrics for the eyes open and eyes closed data.

Fig. 2

Association between age and aperiodic activity. (A) Scatterplot of the aperiodic exponent (upper panel) and offset (lower panel) in relation to age for the eyes open and eyes closed EEG recordings. R-squared (R²) and significance values from the regression analyses are shown (asterisk indicates significance after Bonferroni correction). (B) Correlations between aperiodic activity and age for each of the anterior, central, and posterior electrode clusters. Correlations reached significance across all three locations. (C) EEG electrode cap highlighting the electrodes forming each of the electrode clusters used for the correlations (anterior = yellow, central = blue, posterior = magenta).

Fig. 3

A) Scatter plots of centre frequency for the alpha and beta range in relation to age. Significance values from the regression analyses are shown (asterisk indicates significance after Bonferroni correction). Age was found to significantly predict alpha, but not beta, centre frequency. B) Spectral power plotted in relation to centre frequency for the alpha and beta frequency ranges. Topographic plots show the average power distribution for each of the eyes open and eyes closed recordings for the alpha and beta frequency ranges. Star indicates the electrode used for obtaining the power and centre frequency values used in the analyses (alpha = POz electrode, beta = FCz electrode).

Fig. 4

Scatterplots depicting the association between aperiodic activity and aperiodic-adjusted oscillatory power. Both exponent and offset positively correlated with spectral power in both the alpha and beta bands across the eyes open and eyes closed conditions. Asterisks indicate a significant correlation after Bonferroni correction.