Computational protocol: Neural and Genetic Correlates of the Social Sharing of Happiness

Similar protocols

Protocol publication

[…] We used Statistical Parametric Mapping (SPM) software (SPM12 revision 6225; The Wellcome Department of Cognitive Neurology, London, UK) implemented in MATLAB 2014b (MathWorks Inc., Natick, MA) to analyze the functional images. The first four volumes of each fMRI run were discarded due to unsteady magnetization. After all of the volumes were realigned, differences in slice timing within each image volume were corrected. The reference image was at the center of the volume. The whole-brain 3D MP-RAGE volume was co-registered to the EPI volumes and normalized to the Montréal Neurological Institute T1 image template (ICBM152; Evans et al., ) using a non-linear basis function. Subsequently, normalization parameters were applied to all of the EPI volumes. The normalized EPI images were then spatially smoothed in three dimensions using an 8-mm full-width at half-maximum Gaussian kernel. After the realignment process, we checked the data for head movements >3 mm during the experimental run. Task-related activation was evaluated on a voxel-by-voxel basis using the general linear model at the individual level to generate contrast images. The introduction (2.5 s), imagination (15 s), and rating phases (5 s) were separately modeled using a block design convolved with the canonical hemodynamic response. The introduction and rating phases were considered covariates of no interest in order to partial out their contribution to brain activation in the single participant analyses. Using contrast images related to the imagination phases of the six conditions (positive/absence, neutral/absence, negative/absence, positive/presence, neutral/presence, and negative/presence; Figure ), we conducted a random-effects analysis at the group level (Friston, ) using a full factorial design. This group analysis approach, typically referred to as the Summary Statistics approach, is based on a two-level strategy (Stephan et al., ; Monti, ). We used the following hierarchical two-level linear model:In the above formula, “X” is the single-subject design matrix, “XG” is a group-level matrix, “ε” is the individual-specific error, and “εG” is the group-specific error. The statistical threshold was set at an uncorrected p < 0.001 at the voxel level and a family-wise error (FWE)-corrected p < 0.05 at the cluster level (whole brain). The plot function in SPM12 was used to generate the plots of parameter estimates and 90% confidence intervals (CIs) at the regions of interest (ROIs) (mentalizing/theory-of-mind network), such as voxels [−2, 58, 20] (mPFC), [0, −60, 32] (precuneus), [−56, −10, −20] (temporal pole), and [−48, −64, 32] (TPJ). These parameter estimates were then extracted from MATLAB and used to create bar graphs.Furthermore, neural responses associated with individual happiness ratings were assessed using a parametric modulation analysis, wherein the participant's ratings associated with each event from all conditions were entered as covariates in the individual-level analysis. We then conducted a random-effects analysis at the group level using a one-sample t-test design. The statistical threshold was set at an uncorrected p < 0.001 at the voxel level and an FWE-corrected p < 0.05 at the cluster level (whole brain). A statistical parametric map was masked with an atlas in SPM12 (Labels_neuromorphometrics: anterior cingulate gyrus, temporal pole, superior temporal gyrus, and precuneus). We analyzed the correlations between activities in the mentalizing/theory-of-mind network and changes in happiness rating scores for each condition in the fMRI task. We conducted a group analysis with a multiple regression design using subtracted rating scores for happiness and contrast images (presence – absence) for each valence of the imagined event (positive, neutral, and negative). Based on the result of the full factorial analysis (Table ), ROI analysis was conducted using small volume correction (SVC; a 15-mm radius sphere for the peak coordinates of mPFC, precuneus, TPJ, and temporal pole, which are shown in Table ). The statistical threshold was set at an uncorrected p < 0.001 at the voxel level and an FWE-corrected p < 0.05 at the cluster level (ROI). The plot function in SPM12 was used to generate a plot of the adjusted contrast estimates at voxels [58, 2, −22] (temporal pole) and [52, −62, 32] (TPJ). These parameter estimates were extracted from MATLAB and used to create a scatterplot. [...] We recruited 206 healthy male and female volunteers (age range: 18–23 years, mean age: 19.2 years) following the study's approval by the Ethics Committee of Kobe University (approval number: 2014-10) to study sex differences. All participants provided written informed consent in accordance with the Declaration of Helsinki. All participants were Japanese undergraduate students at Kobe University.We tried to collect as much data as possible given the constraints of our research funding. A statistical power analysis was conducted using G*Power version (Faul et al., ). We assumed that the effect size of this study would be equivalent to that observed in the study by Lebe et al. (), who found a significant association between depressive symptoms and HTR2A genotypes. A priori power analysis was used to estimate the necessary sample size for this study as n = 159 (ANOVA: fixed effects, omnibus, one-way, F-tests; effect size = 0.25; alpha error = 0.05; 1-beta error = 0.8; number of groups = 3).The mean BMI of all participants was 20.6 (range: 15.6–32.7). There were significant differences in age (women: 19.0, men: 19.5, p < 0.01) and BMI (women: 20.2, men: 20.9, p < 0.05) between men and women. Because no participants smoked and only 26 participants consumed alcohol, we discounted these potential confounding factors in subsequent analyses. […]

Pipeline specifications

Software tools SPM, G*Power
Applications Miscellaneous, Magnetic resonance imaging, Functional magnetic resonance imaging
Organisms Homo sapiens
Chemicals Adenine, Guanine, Serotonin