Regulating Emotion to Improve Physical Health
Two hundred and fifty three college students (mean age: 22.6 years; s.d.: 1.01; 135 females) from Beijing, China, participated in this study. Participants had no history of cognitive or neurological disorders (e.g., mental retardation, traumatic brain injury or psychiatric illnesses). The study was approved by the Institutional Review Board of Beijing Normal University. Written informed consent was obtained from all participants before the experiment.
Participants' physical health was assessed with Chinese Constitution Questionnaire (CCQ), which is a national standard inventory on health in China (Zhu et al., 2007). The questionnaire consists of 60 questions on a variety of physical symptoms and health problems, such as, 'I catch cold more easily than others'. Participants were instructed to answer each question, based on their general experiences, on a 5-point Likert scale ranging from 'Never' to 'All the time'. Nine items in the questionnaire are related to psychological problems, such as, 'I get anxious and worried easily'. Given that we focused on examining the association between emotional regulation ability and physical health in this study, and that emotion regulation ability is tightly linked with mental health outcomes such as anxiety and depression (Schutte et al., 2007; Martins et al., 2010), the nine mental health items were excluded in the analyses to obtain a pure measure of physical health not contaminated by mental health. Scores of the remaining 51 items were summed up to yield an overall index of participants' physical health, with higher scores indicating better physical health. Previous studies have shown that the questionnaire has high reliability and validity (Zhu et al., 2007, 2008). To further validate the CCQ in this study, we asked participants to evaluate their physical health status by responding to one statement: 'In general, would you say your health is', with response options ranging from 1 (very poor) to 6 (excellent).
Participants' personality was assessed with the NEO Personality Inventory-Revised (NEO PI-R) (Costa and McCrae, 1992). The NEO PI-R is a self-report inventory consisting of 120 items adhering to the five-factor model of personality. Participants were instructed to rate their agreement to each item on a 5-point Likert scale ranging from 'strongly disagree' to 'strongly agree'. This hierarchically structured inventory allows measurement of personality at both the dimensional and facet level. The five dimensions include neuroticism, extraversion, openness, agreeableness and conscientiousness. Each of the five dimensions has six interrelated but independent subscales measuring personality facets. For example, the neuroticism dimension consists of the facets of anxiety, angry hostility, depression, self-consciousness, impulsiveness and vulnerability. Dimension scores are computed by summing up the corresponding facet scores.
Affective experiences can be measured either as transient fluctuations in mood or as stable individual differences in general affective experience. In our study, participants' general experiences of positive affect (PA) and negative affect (NA) were measured using the Positive and Negative Affect Schedule (PANAS) in the 'general' format (Watson et al., 1988). The PANAS consists of 20 items, half of which represent characteristics of PA, and the remaining items represent that of NA. PA items include 'determined', 'proud' and 'inspire'; NA items include 'jittery', 'guilty' and 'shamed'. Participants indicated to what extent they experienced each affect 'in general', that is, how they feel 'on average' on a 5-point Likert scale ranging from 'very slightly' to 'extremely'.
EI is conceptualized as an array of abilities to perceive, use, understand and regulate emotions that leads to adaptive functioning, and the ability to regulate emotions is one core component in all EI models (e.g., Bar-On, 1997; Mayer and Salovey, 1997). In our study, emotion regulation ability was assessed by the Stress Management Scale of the Emotional Quotient Inventory (EQ-i) (Bar-On, 1997), which is a standardized self-report measure of various aspects of EI. The Stress Management Scale consists of two subscales, each of which composed of nine items. The subscale of stress tolerance assesses the competency to effectively and constructively manage emotions (e.g., 'I know how to deal with upsetting problems'), and the subscale of impulse control assesses the competency to effectively and constructively control emotions (e.g., 'It is a problem controlling my anger') (Bar-On, 1997; Bar-On et al., 2003). Participants were asked to indicate the extent to which each statement accurately described them on a 5-point Likert-type scale, ranging from 'very seldom or not true of me' to 'very often true of me or true of me'.
Scanning was performed on a Siemens 3T Trio scanner (MAGENTOM Trio with a Tim system) with a 12-channel phased-array head coil at the BNU Imaging Center for Brain Research, Beijing, China. MPRAGE, an inversion-prepared gradient echo sequence (bandwidth = 190 Hz/pixel, flip angle = 7°, TR/TE/TI =2.53 s/3.39 ms/1.1 s), was used to acquire 3D T1-weighted whole brain structural images (voxel size 1 mm × 1 mm×1.33 mm, 128 slices).
Voxel-based morphometry (VBM) was used to explore the neural correlates of the behaviorally observed association. VBM provides a quantitative measure of tissue volume per unit volume of spatially normalized image (Ashburner and Friston, 2000). VBM preprocessing was performed with SPM8 (http://www.fil.ion.ucl.ac.uk/spm). Specifically, the images were first segmented into GM, white matter and cerebral spinal fluid, using the unified segmentation approach (Ashburner and Friston, 2005). The segmented GM images were aligned and warped to an iteratively improved study-specific template using nonlinear registration in the DARTEL (Ashburner, 2007). Then, we used the normalization function in the DARTEL to normalize GM images to the Montreal Neurological Institute (MNI) space (2 mm isotropic voxel). The normalized GM images were modulated by multiplying the Jacobian determinants derived from the normalization to preserve the local tissue volumes. The modulated GM images were then smoothed with a Gaussian kernel of 8 mm full width at half maximum. The resulted smoothed modulated normalized GM images were used for further analyses.
We used the Harvard–Oxford subcortical probabilistic structural atlas (Smith et al., 2004) with a probability threshold of 50% to define the left and right amygdalae as regions of interest (ROI). We calculated averaged GM volume across all voxels in each ROI. To determine the neural correlates of behavioral measures, we evaluated Pearson correlations between average GM volume of the left and right amygdalae and the behavioral measures of physical health and emotion regulation. Another way of calculating volumes of the amygdala is based on the automated segmentation procedure (Fischl et al., 2002) provided in Freesurfer 5.1 (http://surfer.nmr.mgh.harvard.edu). The procedure assigns a neuroanatomical label to each voxel in an MR volume based on probabilistic information automatically estimated from a manually labeled training set. The automated segmentations have been found to be statistically indistinguishable from manual labeling (Fischl et al., 2002). Pearson correlations were also calculated between the volumes of bilateral amygdala and behavioral measures. Total brain volume (TBV) was calculated based on all gray and white matter volumes and ventricular volumes to correct for individual differences in whole brain size.
Materials and Methods
Participants
Two hundred and fifty three college students (mean age: 22.6 years; s.d.: 1.01; 135 females) from Beijing, China, participated in this study. Participants had no history of cognitive or neurological disorders (e.g., mental retardation, traumatic brain injury or psychiatric illnesses). The study was approved by the Institutional Review Board of Beijing Normal University. Written informed consent was obtained from all participants before the experiment.
Assessment of Physical Health
Participants' physical health was assessed with Chinese Constitution Questionnaire (CCQ), which is a national standard inventory on health in China (Zhu et al., 2007). The questionnaire consists of 60 questions on a variety of physical symptoms and health problems, such as, 'I catch cold more easily than others'. Participants were instructed to answer each question, based on their general experiences, on a 5-point Likert scale ranging from 'Never' to 'All the time'. Nine items in the questionnaire are related to psychological problems, such as, 'I get anxious and worried easily'. Given that we focused on examining the association between emotional regulation ability and physical health in this study, and that emotion regulation ability is tightly linked with mental health outcomes such as anxiety and depression (Schutte et al., 2007; Martins et al., 2010), the nine mental health items were excluded in the analyses to obtain a pure measure of physical health not contaminated by mental health. Scores of the remaining 51 items were summed up to yield an overall index of participants' physical health, with higher scores indicating better physical health. Previous studies have shown that the questionnaire has high reliability and validity (Zhu et al., 2007, 2008). To further validate the CCQ in this study, we asked participants to evaluate their physical health status by responding to one statement: 'In general, would you say your health is', with response options ranging from 1 (very poor) to 6 (excellent).
Assessment of Personality
Participants' personality was assessed with the NEO Personality Inventory-Revised (NEO PI-R) (Costa and McCrae, 1992). The NEO PI-R is a self-report inventory consisting of 120 items adhering to the five-factor model of personality. Participants were instructed to rate their agreement to each item on a 5-point Likert scale ranging from 'strongly disagree' to 'strongly agree'. This hierarchically structured inventory allows measurement of personality at both the dimensional and facet level. The five dimensions include neuroticism, extraversion, openness, agreeableness and conscientiousness. Each of the five dimensions has six interrelated but independent subscales measuring personality facets. For example, the neuroticism dimension consists of the facets of anxiety, angry hostility, depression, self-consciousness, impulsiveness and vulnerability. Dimension scores are computed by summing up the corresponding facet scores.
Assessment of General Affective Experiences
Affective experiences can be measured either as transient fluctuations in mood or as stable individual differences in general affective experience. In our study, participants' general experiences of positive affect (PA) and negative affect (NA) were measured using the Positive and Negative Affect Schedule (PANAS) in the 'general' format (Watson et al., 1988). The PANAS consists of 20 items, half of which represent characteristics of PA, and the remaining items represent that of NA. PA items include 'determined', 'proud' and 'inspire'; NA items include 'jittery', 'guilty' and 'shamed'. Participants indicated to what extent they experienced each affect 'in general', that is, how they feel 'on average' on a 5-point Likert scale ranging from 'very slightly' to 'extremely'.
Assessment of Emotion Regulation Ability
EI is conceptualized as an array of abilities to perceive, use, understand and regulate emotions that leads to adaptive functioning, and the ability to regulate emotions is one core component in all EI models (e.g., Bar-On, 1997; Mayer and Salovey, 1997). In our study, emotion regulation ability was assessed by the Stress Management Scale of the Emotional Quotient Inventory (EQ-i) (Bar-On, 1997), which is a standardized self-report measure of various aspects of EI. The Stress Management Scale consists of two subscales, each of which composed of nine items. The subscale of stress tolerance assesses the competency to effectively and constructively manage emotions (e.g., 'I know how to deal with upsetting problems'), and the subscale of impulse control assesses the competency to effectively and constructively control emotions (e.g., 'It is a problem controlling my anger') (Bar-On, 1997; Bar-On et al., 2003). Participants were asked to indicate the extent to which each statement accurately described them on a 5-point Likert-type scale, ranging from 'very seldom or not true of me' to 'very often true of me or true of me'.
MRI Data Acquisition and Analysis
Scanning was performed on a Siemens 3T Trio scanner (MAGENTOM Trio with a Tim system) with a 12-channel phased-array head coil at the BNU Imaging Center for Brain Research, Beijing, China. MPRAGE, an inversion-prepared gradient echo sequence (bandwidth = 190 Hz/pixel, flip angle = 7°, TR/TE/TI =2.53 s/3.39 ms/1.1 s), was used to acquire 3D T1-weighted whole brain structural images (voxel size 1 mm × 1 mm×1.33 mm, 128 slices).
Voxel-based morphometry (VBM) was used to explore the neural correlates of the behaviorally observed association. VBM provides a quantitative measure of tissue volume per unit volume of spatially normalized image (Ashburner and Friston, 2000). VBM preprocessing was performed with SPM8 (http://www.fil.ion.ucl.ac.uk/spm). Specifically, the images were first segmented into GM, white matter and cerebral spinal fluid, using the unified segmentation approach (Ashburner and Friston, 2005). The segmented GM images were aligned and warped to an iteratively improved study-specific template using nonlinear registration in the DARTEL (Ashburner, 2007). Then, we used the normalization function in the DARTEL to normalize GM images to the Montreal Neurological Institute (MNI) space (2 mm isotropic voxel). The normalized GM images were modulated by multiplying the Jacobian determinants derived from the normalization to preserve the local tissue volumes. The modulated GM images were then smoothed with a Gaussian kernel of 8 mm full width at half maximum. The resulted smoothed modulated normalized GM images were used for further analyses.
We used the Harvard–Oxford subcortical probabilistic structural atlas (Smith et al., 2004) with a probability threshold of 50% to define the left and right amygdalae as regions of interest (ROI). We calculated averaged GM volume across all voxels in each ROI. To determine the neural correlates of behavioral measures, we evaluated Pearson correlations between average GM volume of the left and right amygdalae and the behavioral measures of physical health and emotion regulation. Another way of calculating volumes of the amygdala is based on the automated segmentation procedure (Fischl et al., 2002) provided in Freesurfer 5.1 (http://surfer.nmr.mgh.harvard.edu). The procedure assigns a neuroanatomical label to each voxel in an MR volume based on probabilistic information automatically estimated from a manually labeled training set. The automated segmentations have been found to be statistically indistinguishable from manual labeling (Fischl et al., 2002). Pearson correlations were also calculated between the volumes of bilateral amygdala and behavioral measures. Total brain volume (TBV) was calculated based on all gray and white matter volumes and ventricular volumes to correct for individual differences in whole brain size.