On the Spectrum

Chapter 7
have been demonstrated previously (Tick, van der Ende et al. 2007). In this study, we applied
two different approaches to study internalizing problems using the CBCL/1½–5. In our first
approach, we used the internalizing broadband scale score (Achenbach and Rescorla 2000).
In our second approach, we used four previously defined classes with distinct patterns of
internalizing problems that were obtained by a latent profile analysis performed on T-scores of
the above mentioned CBCL/1½–5 syndrome scales (Basten, Althoff et al. 2013). This method
classifies children in one of four classes: children with no problem behavior (n=581), children
with predominantly internalizing problems (n=73), children with predominantly externalizing
problems (n=102), and a class with both internalizing and externalizing symptoms (n=38)
(Althoff, Verhulst et al. 2010). Children were assigned to a class based on the most likely class
memberships. This was justified by the high entropy (0.98) of the latent class model. Figure
1 represents the four classes of children and their CBCL/1½–5 internalizing and externalizing
broadband scores.
Neuroimaging
Structural MRI scans were obtained on a 3-T scanner (Discovery MR750, General Electric
Worldwide, Milwaukee) (White, El Marroun et al. 2013). The high-resolution, T1-weighted
image was collected using an inversion recovery fast spoiled gradient recalled sequence with
the following parameters: repetition time = 10.3 msec, echo time = 4.2 msec, inversion time
= 350 msec, number of excitations = 1, flip angle = 16°, readout bandwidth = 20.8 kHz, matrix
256 × 256, imaging acceleration factor of 2, and an isotropic resolution of 0.9 × 0.9 × 0.9
mm3. All children underwent a mock scanning session prior to the scanning procedure.
Image processing
Image quality assurance is described in the supplementary information. Cortical reconstruction
and volumetric segmentation was performed with the FreeSurfer image analysis suite
5.1 (http://surfer.nmr.mgh.harvard.edu/). The technical details of these procedures are
described in a prior publication (Reuter, Schmansky et al. 2012). Briefly, this process included
the removal of non-brain tissue, automated Talairach transformation into standard space,
and segmentation of the subcortical white and gray matter volumetric structures (including
amygdala and the hippocampus), intensity normalization, tessellation of the gray matter
white matter boundary, automated topology correction, and surface deformation. Once the
cortical models were complete, the images underwent surface inflation (Fischl, Sereno et al.
1999), registration to a spherical atlas (Fischl, Sereno et al. 1999), and the parcellation of the
cerebral cortex into units based on gyral and sulcal structure (Desikan, Segonne et al. 2006).
Cortical thickness was calculated as the closest distance from the gray/white boundary to the
gray/cerebrospinal fluid boundary at each vertex on the tessellated surface (Fischl and Dale
2000).
170