The impact of pubertal DHEA on the development of visuospatial oscillatory dynamics

Abstract

The adolescent brain undergoes tremendous structural and functional changes throughout puberty. Previous research has demonstrated that pubertal hormones can modulate sexually dimorphic changes in cortical development, as well as age-related maturation of the neural activity underlying cognitive processes. However, the precise impact of pubertal hormones on these functional changes in the developing human brain remains poorly understood. In the current study, we quantified the neural oscillatory activity serving visuospatial processing using magnetoencephalography, and utilized measures of dehydroepiandrosterone (DHEA) as an index of development during the transition from childhood to adolescence (i.e., puberty). Within a sample of typically developing youth (ages 9-15), a novel association between pubertal DHEA and theta oscillatory activity indicated that less mature children exhibited stronger neural responses in higher-order prefrontal cortices during the visuospatial task. Theta coherence between bilateral prefrontal regions also increased with increasing DHEA, such that network-level theta activity became more distributed with more maturity. Additionally, significant DHEA-by-sex interactions in the gamma range were centered on cortical regions relevant for attention processing. These findings suggest that pubertal DHEA may modulate the development of neural oscillatory activity serving visuospatial processing and attention functions during the pubertal period.

Keywords: hormones; magnetoencephalography; puberty; theta; visuospatial attention.

FIGURE 1

Visuospatial task paradigm (Vis‐attend). Each trial consisted of two periods: (1) fixation lasting an average of 2000 ms (1900–2100 ms variable ISI) with a 400‐ms baseline period, and (2) an 800 ms stimulus‐presentation period with the grid appearing in one of four locations. Participants indicated the lateral location of the stimulus grid relative to the fixation with a button press (left or right)

FIGURE 2

A significant positive correlation between DHEA and age for the whole sample. Male (blue) and female (red) data have been plotted separately only to enhance clarity; this correlation did not significantly differ between the sexes

FIGURE 3

(Left) Time‐frequency spectrograms of significant oscillatory responses during the task. Data are from two occipital sensors (gamma: M2043, alpha/theta: M1922) and one sensor near the left parietal cortex (Beta: M0443) and have been averaged across all participants. Note that spectrograms with the phase‐locked component removed (i.e., induced activity only) are provided in the supplement (Figure S1). Warm colors reflect power increases relative to the baseline, and cool colors represent decreases relative to baseline. Time frequency windows for source imaging (beamforming) were derived from statistical analyses of these sensor‐level spectrograms, which indicated significant bins in theta, alpha, beta, and gamma activity. (Right) Group‐averaged beamformer images of each time‐frequency oscillatory bin of interest across all participants. The theta, alpha, and two gamma oscillatory responses originated in bilateral occipital cortices, whereas beta was centered on the motor cortex and thus not further examined. Color scale bars indicate the strength of responses (pseudo‐t). Warm colors indicate synchronizations; cool colors indicate desynchronizations

FIGURE 4

Correlations between DHEA levels and theta band activity. (a, Left) Whole‐brain correlations showed a negative relationship between DHEA and theta band activity in the right dorsolateral prefrontal cortex, whereby children with higher DHEA levels (i.e., more mature) exhibited less theta activity in this higher‐order, prefrontal region. Brain images are displayed following neurological convention. Color scale bar indicates statistical significance. (Right) Scatterplot shows the correlation between DHEA and theta activity in the peak voxel (age corrected) extracted from the corresponding map to the left. (b, Left) Connectivity analyses revealed increasing theta coherence between two bilateral prefrontal nodes with increasing DHEA levels (i.e., more maturity). (Right) Scatterplot shows increased theta coherence between the bilateral prefrontal regions (age/amplitude corrected) as a function of DHEA. Thus, these data indicate network‐level theta activity increases with increasing DHEA

FIGURE 5

DHEA‐by‐sex interactions in the gamma band. Fisher's r to Z maps showed significant sex differences across cortical regions involved in visuospatial processing. (Top) Males showed stronger associations between DHEA and gamma activity compared to females in all three regions, including the left inferior frontal gyrus, left temporal cortex, and right inferior frontal gyrus. Color scale bar indicates the significance of the DHEA‐by‐sex interaction effects. (Bottom) Scatterplots show correlations between DHEA (corrected for age) and gamma activity in peak voxels extracted from the corresponding map above, which differed between males and females