Developmental trajectories of cortical thickness by functional brain network: The roles of pubertal timing and socioeconomic status

Figures

• 

  Fig. 1. Scatterplot simple linear regression showing mean cortical thickness (mm) by age in years. Regression lines reflect age-related changes in mean cortical thickness for each sex. mm, millimeters.

• 

  Fig. 2. A. Linear cortical thickness trajectories over the developmental age span by Cole-Anticevic functional network. B. Nonlinear cortical thickness trajectories over the developmental age span by Cole-Anticevic functional network. Cole-Anticevic network surface plots shows network location.

• 

  Fig. 3. A, Scatterplot of VIS1 network cortical thickness trajectory. B, First derivative plot of VIS1 network trajectory. There were two periods of significant variance in the rate of change in cortical thickness across the age span: from ages 5.4–13.6 years and from ages 19.4–20.4 years. C, Scatterplot of PMM network cortical thickness trajectory. D, First derivative plot of PMM network trajectory. Cortical thickness significantly decreased over the age span (5.4–21.9 years of age), with varying instantaneous rates of change. E, Scatterplot of ORA network cortical thickness trajectory. F, First derivative plot of ORA network trajectory. Cortical thickness significantly increased from ages 10.1–16.2 years. For first derivative plots, periods of significant change are denoted by colorized line segments. Grey bands show a 95 % point-wise confidence interval, and segments of the spline where the confidence interval does not include zero are denoted by either red (significant decrease) or blue lines (significant increase). VIS1, visual 1; PMM, posterior multimodal; ORA, orbito-affective; f′(x), first derivative. (For interpretation of the references to colour in this figure legend, the…

• 

  Fig. 4. Heatmap displaying Z-score values obtained from Meng test conducted to identify significant differences between the two correlated coefficients. The correlation coefficients are the association between age and mean thickness for a network. VIS2, visual 2; SMN, somatomotor; CON, cingulo-opercular; DAN, dorsal attention; LAN, language; FPN, frontoparietal; AUD, auditory; DMN, default mode; VMM, ventral multimodal; * , significant differences between network correlation coefficients.

• 

  Fig. 5. Cortical surfaces illustrating standardized beta values for age in linear network trajectories. Magnitude of age effect on thickness from steepest to shallowest is: DAN, FPN, DMN, SMN, VIS2, LAN, CON, AUD, and VMM. Functional networks that are nonlinear are shown in grey.

• 

  Fig. 6. Cortical surfaces illustrating Pearson correlation between left and right hemisphere area slopes. Slopes of homologous areas for most networks were significantly correlated, with the exception of the areas that comprise the orbito-affective network which were only moderately correlated.

• 

  Fig. 7. Age by maternal education interaction effects on Cole-Anticevic network and mean cortical thickness. Beta values represent the age trend (i.e., slope) for cortical thickness across maternal education levels. For both mean cortical thickness and cortical thickness in the dorsal attention network, those whose mothers achieved a master’s degree displayed a steeper cortical thinning slope than those whose mothers achieved a bachelor’s degree. * , significant difference of maternal education level in cortical thickness slopes over the age span.