Computational protocol: Psychopathy Moderates the Relationship between Orbitofrontal and Striatal Alterations and Violence: The Investigation of Individuals Accused of Homicide

Similar protocols

Protocol publication

[…] All sMRI data was acquired using a 1.5T GE Signa scanner with a single-shot gradient echo MPRAGE sequence (TR = 25 ms, TE = 6 ms, field of view = 24 cm × 24 cm, matrix = 256 × 256, flip angle = 45°, thickness = 1.2 mm, 124 continuous sagittal slices without gap). CTh and volumetric segmentation were estimated using the FreeSurfer software (FreeSurfer 4.0.5) (; , ), which is a widely documented and automated program for the analysis of brain structure (). Moreover, its validation was previously reaffirmed in studies regarding schizophrenia in contrast with manual measurements ().In short, the FreeSurfer processing streams applied in the present study included three major functions: the removal of non-brain tissue, the segmentation of gray-white matter, and the transformation to a common space (i.e., the Montreal Neurological Institute space), respectively. In terms of volumetric estimation, after the FreeSurfer processing and estimation, GMV of subcortical brain ROIs in the present study including caudate and putamen were further traced manually using BrainSuite (), which would correct small errors. Regarding the estimation of CTh, after normalization of the intensity, gray-white tissue segmentation was used as the starting point for a deformable surface algorithm for the extraction of the pial and gray-white cortical surfaces (). Then, the entire cortex of each participant was visually inspected for accuracies. Manual correction, if necessary, would be carried out using procedures that were established previously (). After spatial normalization of the data, converting it to a cortical surface-based atlas, local CTh was estimated by calculating the shortest distance between a given point on the estimated pial surface and the gray/white matter boundary across the cortical mantle (). Maps were smoothed across the surface and averaged across participants with a 10-mm full-width at half-maximum Gaussian kernel using a non-rigid high-dimensional spherical averaging method. General linear model (GLM) was performed on the effects of each variable on thickness at each vertex for the statistical comparisons of surface maps (; ). [...] The major aim of the present study was to investigate the moderation effect of psychopathic traits in the relationship between structural brain abnormalities (striatum and OFC) and violence. As such, the relationship between these structural brain abnormalities and violence was investigated as a function of psychopathic traits. To be specific, “structural brain abnormalities” refer to the relative differences in terms of the neural correlates between the non-violent group and the violent ones in the present study. By testing the moderation effect, we tested whether the effect of an independent variable (neural correlates) on a dependent variable (violence) varied across the level of a third variable/moderator variable (psychopathy), which interacted with the independent variable (; ). Specifically, we investigated whether the psychopathic personality traits influenced the strength of the relationship between neural correlates and violence. Therefore, the neural correlates were the independent variables and the binary variable (violence vs non-violence) was the outcome variable while the moderator was psychopathy in the present study. The ROIs (striatum and OFC) were chosen based on prior literature as well as the Qdec whole-brain findings in the present study. Based on the GMV/CTh in prior literature and the whole-brain findings devised by the vertex-based analysis using Qdec (a module of FreeSurfer developed to design and execute surface analysis), ROI analyses were performed for the lateral and mOFC (lOFC and mOFC), putamen, and caudate in the present study. The neural correlates of each ROI were computed: (1) total GMV/CTh by the summation of the left and the right hemispheric volumes/CTh; and (2) GMV/CTh by hemisphere. Correlations, between-group t-test and chi-square test were used to assess the association of demographic information with violence, psychopathy and neural correlates (Tables , ). Analyses of the ROI neural correlates were performed using SPSS (Chicago, IL, United States) via employing logistic regressions. Logistic regressions were conducted for the outcome variable (violence), with the psychopathy scores and ROI neural correlates (lOFC, mOFC, caudate and putamen), and the interaction term of the two as the predictors. Significance was set based on a two-tailed alpha level of 0.05 for all tests. IQ and schizophrenia diagnosis were significantly associated with violence in the present study (P < 0.05) (Table ) which is consistent with prior findings (; ). In all analyses, the covariates (age, IQ, sex, SES, schizophrenia diagnosis and whole brain volumes) were controlled for. […]

Pipeline specifications

Software tools FreeSurfer, BrainSuite, SPSS
Applications Miscellaneous, Magnetic resonance imaging
Diseases Brain Diseases