Acceptance is an adaptive emotion regulation strategy characterized by an open and non-judgmental attitude toward mental and sensory experiences. While a few studies have investigated the neural correlates of acceptance in task-based fMRI studies, a gap remains in the scientific literature regarding dispositional use of acceptance, and how this is reflected at a structural level. Therefore, the aim of the present study was to investigate the neural and psychological differences between infrequent acceptance users (i.e., low accepters) and frequent users (i.e., high accepters). Another question was whether high and low accepters differ in personality traits and emotional intelligence.
Method
For the first time, we applied a data fusion unsupervised machine learning approach (mCCA-jICA) to the gray matter (GM) and white matter (WM) of high accepters (n = 50), and low accepters (n = 78) to possibly detect joint GM-WM differences in both modalities.
Results
Our results show that two covarying GM-WM networks separate high from low accepters. The first network showed decreased GM-WM concentration in a fronto-temporal-parietal circuit largely overlapping with the default mode network in high accepters compared to low accepters. The second network showed increased GM-WM concentration in portions of the orbito-frontal, temporal, and parietal areas, which may correspond to a central executive network, also in high accepters compared to low accepters. At the psychological level, the high accepters displayed higher openness to experience compared to low accepters.
Conclusions
Overall, our findings suggest that high accepters compared to low accepters differ in neural and psychological mechanisms. These findings confirm and extend previous studies on the relevance of acceptance as a strategy associated with well-being.
Opmerkingen
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
The ability to regulate emotion is considered to be a core competence for mental health and well-being. Difficulties in regulating emotions have been associated with a broad spectrum of psychological disorders (Dadomo et al., 2018; Frederickson et al., 2018; Kring & Sloan, 2009; Sheppes et al., 2015). Adaptive emotion regulation strategies are considered effective for individuals experiencing emotion dysregulation and are commonly incorporated into various treatments (Leahy et al., 2011; Dadomo et al., 2016, 2018; Grecucci et al., 2017; De Panfilis et al., 2019). Additionally, acceptance is recognized as to be a core construct in third wave behavioral therapies (Hayes, 2004; Kahl et al., 2011) and experiential-dynamic approaches (Frederickson et al., 2018; Grecucci et al., 2020; Messina et al., 2020). In these contexts, acceptance is defined as “the active and aware embrace of private experiences without unnecessary attempts to change their frequency or form” (Hayes et al., 2012). Acceptance is conceptualized as the counter to experiential avoidance, as it involves a mental stance of open curiosity towards ongoing mental and sensory experiences (Goldin et al., 2019; Grecucci et al., 2015). Acceptance is considered an adaptive strategy and is positively associated with wellbeing (Aldao et al., 2010). In one study that examined different adaptive and maladaptive emotion regulation strategies across eight situations, Aldao and Nolen-Hoeksema (2012) found that acceptance was a particularly adaptive strategy. Another study by Dan-Glauser and Gross (2015) compared acceptance to no regulation and found that acceptance led to an increase in positive emotions and emotional expressivity, as well as a decrease in respiration rate. Goldin et al. (2019) also reported that acceptance resulted in a decreased respiration rate and negative emotion, with no significant difference in skin conductance level compared to no regulation.
Previous functional neuroimaging studies have investigated the neural correlates of acceptance through task-based and resting-state fMRI. Kross et al. (2009) reported a reduction in activity in the anterior cingulate cortex (ACC) and medial prefrontal cortex during regulation of negative autobiographical memories by acceptance, compared to no regulation. Another fMRI study by Smoski et al. (2014) showed an increase in BOLD activity in dorsomedial, anterior midcingulate, left lateral prefrontal cortex (PFC), and right dorsolateral PFC regions during acceptance vs no regulation while viewing sad images. Goldin et al. (2019) reported greater BOLD activity in attention and control regions such as the medial prefrontal cortex, dorsolateral prefrontal cortex, and ventrolateral prefrontal cortex, during reading of negative autobiographical scripts in acceptance vs no regulation. An fMRI study conducted by Dixon et al. (2020) depicted involvement of frontoparietal regions including rostrolateral PFC, inferior frontal sulcus, middle frontal gyrus, anterior midcingulate cortex, and posterior middle temporal gyrus during acceptance vs no regulation while reacting to negative self-belief.
To provide coherence to the sparse findings of the literature on acceptance, Messina et al. (2021) conducted a meta-analysis of 13 fMRI studies. Results showed increased activity in executive areas (although only in a subgroup of studies), and more importantly, a major cluster of decreased brain activity located in the posterior cingulate cortex (PCC)/precuneus. These findings challenge a key tenet on a key tenet of neuroscientific models of emotion regulation underlining the necessary involvement of high-level executive processes.
A recent meta-analysis of 42 fMRI studies comparing reappraisal and acceptance confirmed and extended previous results (Monachesi et al., 2023). The results revealed that reappraisal was associated with increased activity in the superior frontal gyrus and left middle frontal gyrus, as well as decreased activity in the left globus pallidus and putamen, whereas acceptance was associated with reduced activity in the posterior cingulate cortex (PCC)/precuneus, a key region of the default mode network (DMN). Moreover, acceptance was associated with increased activity in the ventrolateral prefrontal cortex (VLPFC) and claustrum (Monachesi et al., 2023). Several studies have reported decreased brain activity within DMN regions during acceptance (Dixon et al., 2020; Opialla et al., 2015). The DMN, associated with mind-wandering (Christoff et al., 2009), has been perceived as the opposite of mindfulness (Mrazek et al., 2012). Such findings imply the disruption of ruminative and self-reflective processes independent of executive processes (Ellard et al., 2017; Messina et al., 2021). One intriguing possibility is that such reduction in the activity of the DMN is also associated with a reduction at a structural level (gray matter concentration) in regions ascribed to the DMN. Recent studies have shown that resting state macro-networks (Ghomroudi et al., 2024) can also be found at a structural level using Independent Component Analysis (ICA) based methods (Baggio et al., 2023; Grecucci et al., 2023; Meier et al., 2016; Vanasse et al., 2021).
Personality traits refer to an individual’s relatively stable patterns of thoughts, feelings, and actions (McCrae & Costa Jr., 2003). The Five-Factor Model (FFM) of personality is widely recognized as the most influential and prominent model in psychology (Costa & McCrae, 1989, 1992). According to this model, there are five primary dimensions in which individuals differ in their personality traits: Neuroticism (N): Reflects an individual's tendency to experience negative emotions and psychological distress in response to stressful situations. Extraversion (E): Measures an individual's level of sociability, positive emotionality, and overall activity. Openness to Experience (O): Assesses an individual's curiosity, independent judgment, and willingness to embrace new experiences while considering levels of conservatism. Agreeableness (A): Evaluates an individual's tendencies toward altruism, sympathy, and cooperation in interpersonal relationships. Conscientiousness (C): Assesses the extent to which an individual exhibits self-control in planning, organization, and overall accountability. Several studies have explored the relationship between personality traits and emotion regulation strategies. Barańczuk (2019) explored the relationship between the five-factor model of personality and adaptive as well as maladaptive emotion regulation strategies. The results revealed that higher levels of extraversion, openness to experience, agreeableness, and conscientiousness were positively correlated with the use of adaptive strategies, such as reappraisal, problem-solving, and mindfulness, whereas lower levels of neuroticism were negatively correlated with the use of these adaptive strategies. Conversely, higher levels of neuroticism were positively correlated with maladaptive strategies like avoidance, suppression, worry, and rumination, while lower levels of extraversion, openness to experience, agreeableness, and conscientiousness were negatively correlated with the use of these maladaptive strategies. Additionally, openness to experience showed a positive correlation with reappraisal, problem solving, mindfulness and it was inversely associated with suppression. A meta-analysis of 141 studies examined the relationships between trait mindfulness and personality traits. The findings revealed negative correlations with neuroticism, positive correlations with conscientiousness, and openness to experience (Banfi & Randall, 2022). As far as we know, no study has reported differences between high and low accepters in terms of the personality factors.
Trait Emotional Intelligence (tEI) refers to the ability to comprehend and use emotions intrapersonally and interpersonally. It is defined as a constellation of emotional self-perceptions located at the lower levels of personality hierarchies (Petrides et al., 2007). Some studies have revealed that individuals with high trait emotional intelligence exhibit greater non-judgmental attention to the present moment (Charoensukmongkol, 2014; Wang & Kong, 2014) and engage in less rumination compared to those with low trait emotional intelligence (Ramos et al., 2007; Salguero et al., 2013). A study conducted by Zanella et al. (2022) investigating the connection between tEI and several Emotion Regulation strategies revealed a negative relationship between tEI and certain maladaptive emotion regulation strategies, specifically suppression and self-blame. Notably, no significant association was observed between tEI and acceptance. However, as far as we know, no study clearly explored differences in tEI subscales between high and low accepters.
Previous studies have explored the regulatory nature of acceptance when participants applied this strategy to emotional stimuli during task-based fMRI. However, there is a dearth of research addressing how the dispositional use of emotion regulation strategies is encoded in the brain, how individuals with varying degrees of acceptance ability are sedimented at a structural level, and the potential contributions of white matter, which have been largely overlooked (Ghomroudi et al., 2023; Giuliani et al., 2011b; Kühn et al., 2011; Pappaianni et al., 2018). Furthermore, although some studies have shown gray matter differences related to other emotion regulation strategies (Giuliani et al., 2011b; Pappaianni et al., 2020) the present study aimed to investigate the differences in both gray and white matter in the brains of individuals with high and low levels of acceptance for the first time. Additionally, we explored potential differences in personality traits and emotional intelligence between individuals with high versus low acceptance, in order to overcome the limitations in the existing literature.
The study had a threefold aim: firstly, to identify independent circuits of covarying gray and white matter that differ between individuals with high versus low levels of acceptance, using a data fusion machine learning method. We expected to observe differences in gray and white matter concentration in regions of the default mode network (DMN), particularly the PCC/precuneus which has been linked to reduced mind wandering and rumination among high accepters (Dixon et al., 2020; Messina et al., 2021; Monachesi et al., 2023; Opialla et al., 2015). Alternatively, increased gray and white matter concentration in prefrontal regions involved in inhibiting emotional regions (Morawetz et al., 2017) may be observed. Based on previous studies, we hypothesized that subcortical regions (e.g., insula, cingulate cortex) and frontal regions (e.g., ventrolateral prefrontal cortex) may exhibit increased gray and white matter concentration for high accepters, whereas regions like PCC/precuneus may display reduced concentration in high accepters (Messina et al., 2021; Monachesi et al., 2023). Secondly, the present study aimed to identify possible differences between high and low accepters in terms of personality traits. Drawing from a previous study (Barańczuk, 2019), we hypothesized that high acceptance is positively associated with traits such as openness to experience, conscientiousness and negatively correlated with traits such as neuroticism. Finally, the study aimed to explore potential differences in emotional intelligence (EI) between high and low accepters. Building on previous studies, we hypothesized that high acceptance is positively associated with emotional intelligence (Chu, 2010; Schutte & Malouff, 2011). One possibility is that high accepters differ from low accepters in several dimensions associated with personality traits and emotional intelligence.
Of note, from a methodological point of view, our study departed from the previous attempts to characterize emotion regulation abilities which largely relied on univariate methods, by using a multivariate unsupervised machine learning approach. Notably, the present study used multimodal canonical correlation analysis (mCCA) in combination with joint independent component analysis (jICA) (Sui et al., 2012), a multivariate machine learning technique. Previous studies have applied univariate statistical techniques, such as voxel-based morphometry, which do not account for voxel dependencies across the entire brain (Giuliani et al., 2011a; Hermann et al., 2013; Mak et al., 2009).Conversely, multivariate approaches such as Independent Component Analysis (ICA) based methods (Grecucci et al., 2023; Norman et al., 2006; Sorella et al., 2019; Xu et al., 2009) and mCCA + jICA, consider the statistical dependencies among voxels and can identify complex and sparse patterns throughout the entire brain. As a result, these approaches do not require defining a priori regions of interest (ROIs) that are typically based on brain atlases derived from histological properties and biologically implausible parcellations of the brain. The mCCA + jICA merges two modalities, such as GM and WM, to identify the correlations between them using mCCA and separate the covariance matrix into independent networks of covarying GM-WM using jICA. The fusion of these two methods allows for a multimodal fusion (MMF) approach to identify the unique and shared variance associated with each modality in relation to cognitive functioning in individuals. This method has been successfully implemented in several studies. For example, Liang et al. (2021) applied mCCA-jICA to investigate the covariation of gray and white matter concentration in cognitive decline and mild cognitive impairment, Baggio et al. (2023) used this method to investigate individual differences in anxiety trait, and Kim et al. (2015)employed mCCA-jICA to explore gray and white matter networks that contribute to structural alterations in the brains of patients with OCD (Obsessive Compulsive Disorder). The multi-modal fusion approach provides a reliable interrelationship of changes in each modality, yielding more integrated information about the brain. Multi-modal fusion approaches are particularly suitable for analyzing complicated and weak effects that may be undetectable in high-dimensional data sets, while also exhibiting robustness to noise (Calhoun & Sui, 2016).
Method
Participants
T1 structural brain images, CERQ, TEIQue-SF questionnaire scores, and NEO-Five-Factor Inventory (NEO-FFI) of one group of high accepters (n = 58, 16 Females, mean acceptance score M = 9.68 ± 1.21,mean age M = 29.48 ± 12.2 and average 12.5 ± 0.91 years of education), and another group of low accepters (n = 70, 20 Females, mean acceptance score 4.84 ± 1.71, mean age M = 29.92 ± 12.7 years and average M = 12.5 ± 0.91 years of education) of German native speakers were included in this study. The data were selected from “Leipzig study for mind–body-emotion interactions” (OpenNeuro database, accession number ds000221) (LEMON). Data collection was conducted at the Max Planck Institute for Human Cognitive and Brain Sciences (MPI CBS) in Leipzig. Data collection proceeded in compliance with the Declaration of Helsinki, and the University of Leipzig's medical faculty ethics committee approved the study protocol (reference number 154/13-ff) (Babayan et al., 2019). The exclusion criteria for data collection were as follows: no cardiovascular disease, history of psychiatric diseases, history of neurological disorders, history of malignant diseases or intake of centrally active medication, beta- and alpha-blocker, cortisol, any chemotherapeutic or psychopharmacological medication. We also excluded participants who reported drug or excessive alcohol use.
A total of 128 participants (36 females) were included in the study after excluding 7 participants due to corrupted data. Participants provided written informed consent and agreed to anonymous data sharing prior to data collection. They received compensation for participating in the study after the completion of all assessments.
Measures
The study used three questionnaires to assess various constructs. The first questionnaire utilized was the Cognitive Emotion Regulation Questionnaire (CERQ) German version (Garnefski et al., 2001; Loch et al., 2011). The CERQ measures nine cognitive coping strategies, including self-blame, acceptance, rumination, positive refocusing, refocus on planning, positive reappraisal, putting into perspective, catastrophizing, and blaming others. The German version of the questionnaire consists of 36 items, with each strategy measured through four questions. Participants responded using a 5-point Likert scale, ranging from 1 (almost never) to 5 (almost always). For the purposes of this study, the Acceptance scale was specifically focused on to differentiate the two groups and investigate neural and psychological differences between high and low accepters. The second questionnaire employed was the German adaptation of the NEO-Five-Factor Inventory (Borkenau & Ostendorf, 2008) to evaluate the Big Five of Personality Inventory (NEO-FFI; Costa & McCrae, 1992). This questionnaire consists of 60 items, which are categorized into five personality factors: Neuroticism, Extraversion, Openness to experience, Agreeableness, and Conscientiousness. Responses are provided on a 5-point Likert scale, ranging from 0 (strong denial) to 4 (strong approval). Lastly, the German adaptation (TEIQue-SF, Freudenthaler et al., 2008) of the Trait Emotional Intelligence Questionnaire-Short Form (TEIQue-SF; Petrides & Furnham, 2006) was used. This questionnaire comprises four subscales: Well-Being, Self-Control, Emotionality, and Sociability. It includes a total of 30 items, with two items from each of the 15 facets of the TEIQue. To investigate whether the frequency of using acceptance leaves a structural trace in the brain, participants were divided into two subgroups based on the median value (−0.0136) of the z-scores of the Acceptance subscale of the CERQ questionnaire. Individuals with scores below the median were assigned to the low accepters group (70 individuals) and those with scores above the median were assigned to high accepters group (58 individuals).
Procedure
The T1-weighted structural images were acquired using a 3 Tesla scanner (MAGNETOM Verio, Siemens Healthcare GmbH, Erlangen, Germany) equipped with a 32-channel head coil. The scanner was not subject to any major maintenance or updates during the experiment and the data quality remained stable throughout the study.
First, structural MRI scans were initially inspected manually to assess their quality, ensuring the exclusion of any possible artifacts. The structural images were then reoriented manually to the anterior commissure as the reference point. Next, the data was pre-processed using Computational Anatomy Toolbox (CAT12, http://www.Neuro.uni-jena.de/cat/), a toolbox for statistical Parametric Mapping (SPM12) in MATLAB environment (The Mathworks, Natick, MA, USA).
The CAT12 toolbox was used to segment the images into gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF). To enhance registration accuracy, Diffeomorphic Anatomical Registration of GM images was conducted using Exponential Lie Algebra (DARTEL) tools for SPM12 instead of traditional whole brain registration (Grecucci et al., 2016; Pappaianni et al., 2018; Yassa & Stark, 2009). Finally, the DARTEL images were normalized to the MNI-152 standard space and smoothed using a 12-mm full-width at half-maximum (FWHM) Gaussian kernel [12, 12, 12].
Data Analyses
The fusion analysis of mCCA-jICA was performed using the fusion ICA Toolbox (FITv2.0d, http://mialab.mrn.org/software/fit/) (Fig. 1). First, GM and WM features were extracted, which resulted in two feature matrices with rows representing the number of participants and columns representing the number of voxels. These matrices were then normalized to z-scores. After applying the default parameters, a total of 12 independent components were extracted. The number of components was estimated for both modalities using the minimum description length criterion (Li et al., 2007), Finally, singular value decomposition (SVD) was used to reduce dimensionality of feature matrices.
Fig. 1
mCCA + jICA method: First, the structural T1 images were preprocessed then the GM and WM features were extracted. Next, the features were reorganized into two matrices and the dimensionality was reduced. Furthermore, mCCA was applied to the feature matrices then canonical variants and the associated components were computed. Finally, jICA was applied to compute the maximized joint-independent component matrices and the relative loading coefficients mixing matrix
×
To find which IC correctly classifies high and low accepters, backward logistic regression was performed in (JASP Team, 2021) backward regression is a widely used wrapper method for feature selection. It involves iteratively removing the least statistically significant independent variable from an initial model containing all independent variables, until a model with optimal performance is obtained (Jollans et al., 2019). Backward logistic regression is a specific type of backward regression that is commonly used when the dependent variable is binary. Backward regression is especially useful when the number of independent variables is not large (Smith, 2018). In this study high and low accepter scores were entered as dependent variable and age, gender and twelve loading coefficients were entered in as independent variables. A power analysis was conducted in R prior to the regression analysis to estimate the effect size and confirm adequate statistical power. Using a significance level of α = 0.05, a power of 0.90, and 14 predictors, the analysis yielded an effect size of d = 0.19.
Statistical analysis of behavioral data was conducted using JASP Team (2021). JASP (Version 0.16). Welch's t-test was conducted in order to find differences in high and low accepters in four subscales of five personality factors: Neuroticism, Extraversion, Openness to experience, Agreeableness, and Conscientiousness, and emotional intelligence: Well-Being, Self-Control, Emotionality, and Sociability and. To control for Type I error, a false discovery rate (FDR) correction was applied to the p-values.
Results
Covarying GM-WM Networks that Separate High from Low Accepters
mCAA and jICA were applied to structural data of 128 individuals and returned 12 independent circuits of covarying gray and white matter components. The number of components was estimated by the software according to the Information theoretic criteria. The values obtained from these networks reflect changes in gray and white matter concentration. Positive values indicate an increase in concentration, while negative values indicate a decrease. Covariation between gray and white matter components suggests a similar pattern of concentration changes in both types of matter (Fig. 2).
Fig. 2
Independent components from 1 to 12. ICA was able to decompose the brain into 12 covarying gray and white matter networks. The colors represent differences in gray/white matter concentration. Red indicates lower concentration in the areas depicted and yellow indicates higher concentration
×
Backward Logistic Regression Analysis
The backward logistic regression to characterize high versus low accepters returned as significant the IC7 (β = −9.15, SE = 4.14, Wald = 5.35, p = 0.027), which showed decreased GM-WM concentration in fronto-temporal-parietal regions for high accepters compared to low accepters, and IC10 (β = 21.71, SE = 9.17, Wald = 5.75, p = 0.018), which showed increased GM-WM concentration in the parts of orbito-frontal, temporal and parietal cortices in high accepters compared to low accepters (Tables 1 and 2; Figs. 3 and 4).
Table 1
Result from backward logistic regression analysis with winning models IC7 and IC10 for high and low accepters
Source
β
SE
Wald χ2
p
OR
95% CI
IC7
−9.15
4.14
5.37
0.02
1.055 × 10–4
[0, 0.35]
IC10
21.71
9.17
5.75
0.01
2.702 × 10+9
[2.77, 22.47]
OR Odds ratio; CI Confident interval; SE Standard error
Table 2
Gray and white independent component 7, 10
Network
Area
Brodmann area
Volume
Random effects: max value (x, y, z)
GM-WM (cc)
IC7 GM
Cerebellar Tonsil
*
0.4/3.8
4.9 (−34, −38, −35)/8.4 (46, −48, −35)
*
*
0.1/0.4
3.7 (−10, 4, −7)/6.1 (1, 19, 53)
Paracentral Lobule
4, 5, 6, 31
2.4/2.2
6.8 (−1, −30, 61)/6.5 (4, −30, 61)
Medial Frontal Gyrus
6, 8
1.6/3.0
6.7 (−3, −26, 61)/6.6 (4, −25, 61)
Declive
*
2.5/0.0
6.7 (−18, −82, −17)/−999.0 (0, 0, 0)
Middle Frontal Gyrus
6, 8, 9, 10
3.2/1.9
5.3 (−31, 38, 26)/6.0 (25, 34, 34)
Superior Frontal Gyrus
6, 8, 9, 10
1.7/3.4
5.9 (−28, 51, −1)/6.0 (13, 9, 59)
Sub-Gyral
6
3.2/3.2
5.3 (−42, 23, 18)/6.0 (30, −76, 23)
Lentiform Nucleus
*
1.0/2.2
4.8 (−16, 3, −3)/5.9 (21, 0, −1)
Posterior Cingulate
18, 29, 30
0.7/1.7
5.1 (−19, −58, 10)/5.7 (18, −66, 12)
Culmen
*
0.5/1.1
4.1 (−12, −63, −3)/5.7 (46, −48, −30)
Precentral Gyrus
4
0.6/1.7
4.5 (−9, −23, 63)/5.6 (21, −23, 59)
Precuneus
7, 19
1.2/0.5
5.5 (−25, −59, 42)/4.6 (27, −76, 20)
Extra-Nuclear
*
0.4/1.1
4.6 (−19, −53, 10)/5.4 (22, −55, 8)
Inferior Parietal Lobule
40
0.1/1.3
4.2 (−31, −56, 43)/5.3 (53, −41, 27)
Lingual Gyrus
17, 18, 19
2.5/0.8
5.3 (−24, −70, −4)/5.0 (15, −51, 5)
Superior Parietal Lobule
7
0.8/0.0
5.3 (−24, −60, 46)/−999.0 (0, 0, 0)
Cuneus
17, 18, 19, 23, 30
0.9/1.8
4.9 (−12, −71, 10)/5.1 (28, −79, 26)
Parahippocampal Gyrus
30
0.1/0.6
3.6 (−24, −10, −16)/5.1 (12, −48, 4)
Uvula
*
0.4/0.0
5.0 (−19, −78, −23)/−999.0 (0, 0, 0)
Middle Temporal Gyrus
21, 38
0.3/0.4
4.0 (−48, 8, −18)/5.0 (33, −76, 20)
Postcentral Gyrus
3, 4, 5
1.2/0.7
4.8 (−21, −32, 60)/5.0 (22, −26, 62)
Fusiform Gyrus
18, 19
0.6/0.0
4.8 (−21, −88, −15)/−999.0 (0, 0, 0)
Middle Occipital Gyrus
18, 19
0.0/0.4
−999.0 (0, 0, 0)/4.8 (30, −79, 21)
Inferior Frontal Gyrus
46
0.3/0.4
4.2 (−45, 23, 15)/4.8 (48, 21, 17)
Supramarginal Gyrus
*
0.1/0.4
3.7 (−50, −51, 25)/4.7 (53, −41, 31)
Superior Temporal Gyrus
13, 22, 38
1.5/1.2
4.6 (−45, 11, −19)/4.5 (43, 9, −24)
Tuber
*
0.0/0.2
−999.0 (0, 0, 0)/4.6 (46, −54, −30)
Superior Occipital Gyrus
*
0.0/0.1
−999.0 (0, 0, 0)/4.6 (33, −76, 26)
Cingulate Gyrus
24, 32
0.2/1.0
4.2 (−1, −8, 47)/4.4 (4, 25, 36)
Lateral Ventricle
*
0.0/0.2
−999.0 (0, 0, 0)/4.2 (28, −55, 8)
Anterior Cingulate
32
0.3/0.0
4.0 (−1, 30, 25)/−999.0 (0, 0, 0)
Angular Gyrus
*
0.1/0.0
3.9 (−31, −58, 37)/−999.0 (0, 0, 0)
Inferior Semi-Lunar Lobule
*
0.0/0.1
−999.0 (0, 0, 0)/3.8 (25, −69, −36)
Pyramis
*
0.0/0.1
−999.0 (0, 0, 0)/3.7 (21, −72, −32)
Inferior Temporal Gyrus
*
0.0/0.1
−999.0 (0, 0, 0)/3.6 (59, −15, −16)
Subcallosal Gyrus
34
0.0/0.1
−999.0 (0, 0, 0)/3.6 (15, 4, −13)
Inferior Occipital Gyrus
18
0.1/0.0
3.6 (−21, −88, −9)/−999.0 (0, 0, 0)
IC7 WM
Extra-Nuclear
*
1.5/1.0
6.9 (−19, −37, 21)/5.4 (21, −35, 21)
Precuneus
7, 19, 39
3.1/2.8
5.7 (−27, −50, 52)/6.7 (13, −62, 51)
Inferior Parietal Lobule
7, 39, 40
4.4/2.4
6.5 (−46, −42, 42)/5.4 (55, −35, 27)
Lateral Ventricle
*
1.0/0.5
6.5 (−16, −34, 21)/5.1 (19, −31, 22)
Middle Temporal Gyrus
21, 39
1.6/2.5
6.4 (−49, 2, −25)/4.9 (42, 4, −30)
Middle Occipital Gyrus
18, 19
0.1/2.1
3.9 (−42, −77, 11)/5.8 (39, −76, 1)
Supramarginal Gyrus
40
0.1/0.8
3.7 (−46, −42, 37)/5.7 (53, −38, 31)
Medial Frontal Gyrus
10
0.9/2.8
5.6 (−12, 57, −4)/5.5 (13, 57, 8)
Superior Frontal Gyrus
9, 10, 11
1.5/1.6
5.5 (−13, 53, −11)/5.1 (10, 56, −7)
Superior Parietal Lobule
5, 7
1.0/0.5
5.1 (−30, −51, 55)/5.5 (13, −64, 54)
Cuneus
7, 17, 18, 19
1.0/0.6
5.5 (−19, −80, 32)/4.6 (25, −87, 11)
Cerebellar Tonsil
*
0.0/3.1
−999.0 (0, 0, 0)/5.4 (28, −51, −37)
Middle Frontal Gyrus
9, 10
1.0/1.4
4.9 (−31, 49, −5)/5.2 (34, 49, −4)
Sub-Gyral
40
1.9/1.0
5.1 (−25, −45, 52)/5.0 (36, −74, −1)
Insula
13
0.0/0.3
−999.0 (0, 0, 0)/4.9 (58, −33, 18)
Postcentral Gyrus
2, 3, 40
0.2/1.3
4.3 (−46, −29, 37)/4.9 (46, −26, 39)
Inferior Temporal Gyrus
20, 21
1.2/0.4
4.9 (−49, −1, −28)/4.2 (56, −13, −17)
Superior Temporal Gyrus
22, 38, 42
0.1/1.0
3.8 (−49, 6, −22)/4.7 (56, −33, 14)
Fusiform Gyrus
20
0.5/0.2
4.7 (−49, −4, −25)/4.3 (52, −3, −25)
*
*
0.0/0.0
−999.0 (0, 0, 0)/−999.0 (0, 0, 0)
Angular Gyrus
39
0.6/0.0
4.6 (−43, −59, 36)/−999.0 (0, 0, 0)
Cingulate Gyrus
*
0.3/0.0
4.6 (−16, −34, 27)/−999.0 (0, 0, 0)
Precentral Gyrus
44
0.1/0.4
3.6 (−49, 18, 7)/4.3 (55, −19, 35)
Paracentral Lobule
5
0.3/0.0
4.1 (−19, −41, 50)/−999.0 (0, 0, 0)
Inferior Semi-Lunar Lobule
*
0.0/0.3
−999.0 (0, 0, 0)/4.1 (31, −62, −39)
Inferior Occipital Gyrus
19
0.0/0.2
−999.0 (0, 0, 0)/4.0 (36, −77, −4)
Caudate
*
0.1/0.1
3.6 (−9, 17, 12)/4.0 (19, −28, 19)
Inferior Frontal Gyrus
45
0.1/0.0
4.0 (−49, 20, 11)/−999.0 (0, 0, 0)
IC10 GM
*
*
0.1/0.3
5.5 (−3, 12, −22)/6.4 (3, 21, −24)
Rectal Gyrus
11
0.4/0.4
5.3 (−6, 18, −25)/5.8 (3, 23, −26)
Caudate
*
1.3/0.3
4.7 (−12, 14, 9)/3.9 (15, 15, 8)
Sub-Gyral
*
0.2/0.3
3.6 (−21, 25, 37)/4.4 (42, 19, 23)
Superior Frontal Gyrus
6, 8, 9
1.0/1.2
4.3 (−22, 44, 32)/4.3 (10, 10, 58)
Medial Frontal Gyrus
6
0.8/0.4
4.1 (−3, −22, 62)/3.9 (1, −11, 50)
Middle Frontal Gyrus
8
0.4/0.2
4.1 (−24, 25, 40)/3.7 (45, 22, 22)
Tuber
*
0.0/0.3
−999.0 (0, 0, 0)/4.1 (36, −58, −30)
Lentiform Nucleus
*
0.4/0.6
4.0 (−15, 3, −4)/4.0 (18, 3, −3)
Culmen
*
0.0/0.6
−999.0 (0, 0, 0)/4.0 (24, −35, −22)
Extra-Nuclear
*
0.3/0.1
4.0 (−15, 11, 9)/3.7 (33, 20, −2)
Inferior Parietal Lobule
*
0.0/0.1
−999.0 (0, 0, 0)/4.0 (39, −42, 41)
Insula
13
0.0/0.1
−999.0 (0, 0, 0)/3.9 (33, 23, 0)
Middle Temporal Gyrus
*
0.0/0.1
−999.0 (0, 0, 0)/3.9 (43, −55, 12)
Precuneus
7
0.3/0.0
3.8 (−22, −65, 36)/−999.0 (0, 0, 0)
Cerebellar Tonsil
*
0.0/0.1
−999.0 (0, 0, 0)/3.7 (33, −58, −32)
Paracentral Lobule
*
0.1/0.0
3.6 (−1, −29, 58)/−999.0 (0, 0, 0)
Inferior Frontal Gyrus
*
0.1/0.0
3.5 (−43, 34, 12)/−999.0 (0, 0, 0)
IC10 WM
Precentral Gyrus
6, 9, 44
1.2/0.9
11.0 (−37, 2, 36)/6.6 (45, 5, 36)
Sub-Gyral
31
11.9/7.7
9.8 (−34, 2, 33)/6.8 (42, 9, 15)
Middle Frontal Gyrus
6, 8, 9, 10
3.0/1.7
9.2 (−40, 2, 39)/5.2 (48, 8, 37)
Inferior Frontal Gyrus
6, 9, 47
1.0/0.8
9.1 (−37, −1, 33)/5.0 (37, 3, 33)
Precuneus
7, 19, 31, 39
1.0/2.1
7.0 (−33, −64, 36)/7.0 (25, −49, 43)
Superior Parietal Lobule
7
0.7/1.0
4.8 (−33, −66, 43)/6.9 (28, −50, 45)
Medial Frontal Gyrus
6, 8, 9, 32
3.6/0.6
6.5 (−25, 43, 8)/4.2 (22, 43, 17)
Supramarginal Gyrus
*
1.4/0.1
6.5 (−43, −52, 29)/3.5 (46, −52, 34)
Superior Frontal Gyrus
6, 9, 10
2.8/2.7
6.1 (−15, 13, 52)/6.4 (21, 55, 15)
Superior Temporal Gyrus
22, 39
1.2/0.1
6.2 (−43, −51, 25)/3.6 (55, −37, 17)
Insula
13
0.0/1.7
−999.0 (0, 0, 0)/6.1 (37, 9, 15)
Inferior Parietal Lobule
40
0.3/1.1
5.1 (−30, −53, 47)/5.9 (31, −53, 45)
Cingulate Gyrus
24, 31, 32
3.4/0.4
5.9 (−12, 8, 45)/4.7 (19, −43, 37)
Middle Temporal Gyrus
20, 39
1.0/0.7
5.8 (−33, −63, 27)/5.0 (55, −33, −12)
Anterior Cingulate
9, 32
0.0/0.8
−999.0 (0, 0, 0)/5.7 (15, 36, 12)
Cerebellar Tonsil
*
1.8/0.8
5.7 (−25, −45, −32)/5.1 (22, −45, −32)
*
*
0.4/0.4
4.8 (−22, −42, −29)/4.5 (21, −42, −29)
Cuneus
17, 18, 30
2.4/1.8
5.2 (−9, −76, 23)/5.5 (19, −73, 6)
Middle Occipital Gyrus
18
1.2/0.4
5.4 (−16, −90, 14)/3.9 (16, −88, 14)
Lingual Gyrus
18
0.1/1.4
4.1 (−31, −72, −4)/5.0 (10, −85, −3)
Paracentral Lobule
5
0.5/0.1
4.9 (−16, −35, 50)/3.7 (18, −42, 57)
Angular Gyrus
*
0.3/0.0
4.8 (−34, −58, 35)/−999.0 (0, 0, 0)
Posterior Cingulate
30, 31
0.6/0.1
4.7 (−13, −65, 14)/4.3 (25, −67, 6)
Fusiform Gyrus
20, 37
0.3/0.1
4.6 (−30, −53, −9)/3.5 (49, −36, −21)
Inferior Occipital Gyrus
*
0.1/0.0
4.5 (−34, −74, −4)/−999.0 (0, 0, 0)
Extra-Nuclear
*
0.3/0.6
3.8 (−22, 18, −10)/4.5 (33, 10, 16)
Postcentral Gyrus
3
0.1/0.3
3.5 (−30, −21, 43)/4.1 (18, −38, 59)
Lateral Ventricle
*
0.2/0.3
4.1 (−4, 19, 0)/3.8 (28, −64, 6)
Rectal Gyrus
11
0.0/0.1
−999.0 (0, 0, 0)/4.0 (6, 21, −25)
Inferior Temporal Gyrus
20
0.1/0.3
3.6 (−39, −70, 0)/3.9 (52, −33, −15)
Declive of Vermis
*
0.1/0.0
3.9 (−1, −69, −15)/−999.0 (0, 0, 0)
Culmen
*
0.1/0.0
3.8 (−33, −44, −29)/−999.0 (0, 0, 0)
Caudate
*
0.1/0.0
3.6 (−7, 19, −2)/−999.0 (0, 0, 0)
Talairach labels of regions of interest, Brodmann area, volume (expressed in cc) and max values coordinates are shown
Fig. 3
Brain plots of Network 1. Top. Brain plot reconstruction of IC7 gray matter concentration of high accepters. Central: gray and white matter concentration of IC7. Bottom: loading coefficients distribution between low and high accepters
Fig. 4
Brain plots of Network 2. Top. Brain plot reconstruction of IC10 gray matter concentration of high accepters. Central. White matter concentration of IC10. Bottom: loading coefficients distribution between low and high accepters
×
×
Behavioral Results
A Welch's t-test was conducted to examine the difference between low and high accepters in terms of emotional intelligence and five personality factors. The result revealed a significant difference between low and high accepters, with FFM of personality—openness to experience higher in high accepters t(122.92) = −1.76, p = 0.040, pFDR = 0.040- (Table 3; Figs. 5 and 6).
Table 3
Result from independent samples t-test analysis
Group
n
Mean
SD
Age (Ave)
Low Accepters
70
29.92
12.71
t(126) = 0.20, p = 0.58, d = −0.03
High Accepters
58
29.48
12.21
gender (F = 1, M = 2)
Low Accepters
70
F = 20
0.45
t(126) = −0.12, p = 0.45, d = −0.02
High Accepters
58
F = 16
0.45
NEOFFI_ Neuroticism
Low Accepters
70
1,4
0,6
t(124.44) = −1.31, p = 0.09, d = 0.23
High Accepters
58
1,54
0,55
NEOFFI_ Extraversion
Low Accepters
70
2,56
0,53
t(118.57) = 1.40, p = 0.91, d = 0.25
High Accepters
58
2,4
0,51
NEOFFI_ Openness to Experiences
Low Accepters
70
2,61
0,52
t(122.92) = −1.76, p = 0.040*, d = −0.31
High Accepters
58
2,76
0,5
NEOFFI_ Agreeableness
Low Accepters
70
2,77
0,41
t(116.10) = −1.15, p = 0.87, d = −0.20
High Accepters
58
2,8
0,45
NEOFFI_ Conscientiousness
Low Accepters
70
2,74
0,54
t(114.69) = 1.62, p = 0.94, d = 0.29
High Accepters
58
2,53
0,63
TeiQueSF self_control
Low Accepters
70
5.09
0.81
t(124.48) = 0.19, p = 0.57, d = 0.03
High Accepters
58
5.06
0.74
TeiQueSF emotionality
Low Accepters
70
5.17
0.76
t(114.68) = 1.34, p = 0.90, d = 0.24
High Accepters
58
4.97
0.86
TeiQueSF sociability
Low Accepters
70
5.11
0.63
t(97.59) = 1.97, p = 0.97, d = 0.35
High Accepters
58
4.82
0.93
TeiQueSF well_being
Low Accepters
70
5.79
0.80
t(115.73) = 0.82, p = 0.79, d = 0.14
High Accepters
58
5.66
0.89
TeiQueSF_total
Low Accepters
70
156
15
t(106.61) = 1.43, p = 0.92, d = 0.25
High Accepters
58
151.81
19.58
* for p < 0.05, ** for p < 0.01, and *** for p < 0.001)
Fig. 5
Personality subscales for low vs high accepters
Fig. 6
Emotional intelligence subscales for low vs high accepters. Note: * for p < 0.05
×
×
Discussion
The aim of this study was to identify joint gray and white matter differences that characterize high accepters compared to low accepters. To this aim, a novel data fusion unsupervised machine learning approach known as mCCA + jICA was used. mCCA + jICA detected two brain networks which, significantly differentiate high accepters and low accepters. The first network included co-altered gray and white matter concentration in the default system, in the salience network and in subcortical areas (basal ganglia and hippocampus), with reduced matter concentration for high accepters. The second network included co-altered gray and white matter concentrations in several executive areas in the prefrontal (dorsomedial prefrontal and orbito-frontal cortices) and parietal cortex, with higher gray matter concentration for high accepters. In what follows, we discuss more in detail the role of these networks.
Network 1 included several areas of the default system (posterior and anterior midline structures, anterior temporal areas, angular gyrus), the dorsal anterior cingulate cortex (dACC) and right insula, and a series of subcortical structures resulting in joint, co-altered components that differ between high and low accepters, with lower matter concentration in high accepters. The implication of the default system in acceptance is in line with previous functional MRI studies which have reported decreased activity in the default system and decreased connectivity within the default system areas. These changes have been associated with individual differences in acceptance-based meditation experience (Brewer et al., 2011) and with acceptance considered as a component of trait mindfulness (Wang et al., 2014; Harrison et al., 2019). In task-related studies of acceptance, it has been suggested that reduced DMN activity, especially in the posterior cingulate and adjacent areas, may reflect the interruption of ruminative processing that characterizes acceptance (Ellard et al., 2017; Messina et al., 2021). On the contrary, hyper-activation of the DMN characterizes rumination and related dysregulated states (Sheline et al., 2009; Messina et al., 2016; Whitfield-Gabrieli & Ford, 2012; Buckner & DiNicola, 2019). According to semantic models of emotion regulation (Messina et al., 2015, 2020; Viviani, 2013), the DMN may contribute to emotion regulation by modulating the semantic processing of emotional stimuli, with automatic activation of internal semantic representations corresponding to higher DMN activation. In line with such model, decreased gray and white matter concentration in association to higher use of acceptance may correspond to a more detached attitude toward emotionally relevant semantic representations, contrasting with a more engaged attitude toward internally generated contents.
The concept of a detached attitude toward internally generated semantic representations aligns with the co-alteration observed in the basal ganglia (lentiform nucleus) and in the hippocampus in the present study, as well as in a previous study on trait mindfulness (Taren et al., 2013). As part of the basal ganglia, the role of lentiform nucleus in the reward response has been well established (Schultz, 2016). Also, the joint activation of basal ganglia and hippocampus structures seems to contribute to reward mechanisms by forming and storing memories of events and places of rewarding experiences (Singer & Frank, 2009; LeGates et al., 2018). The finding of reduced gray matter in reward-related subcortical regions in high accepters is in line with previous findings that reported less susceptibility to extrinsic incentive in expert meditators compared to non-meditators (Brown et al., 2013; Kirk et al., 2015), reflecting a sort of inner peace state in meditators. Also, the right insula and the ACC may contribute to the achievement of this accepting attitude toward internal arousing representations. These areas are key nodes of the salience network, which integrates internal and external information to guide behavior (Seeley, 2019). The salience network is responsible for switching between the DMN and the central executive network, signalling the DMN to reduce its activity when attention is shifted from internal to external focus (Goulden et al., 2014; Jilka et al., 2014). Taken together, these results account for a difference between high accepters and low accepters in the management of internally generated arousing contents.
The second network detected in the present study was composed of a series of prefrontal and superior parietal areas, with a possible overlap with the dorsal fronto-parietal network associated with top-down attention processes (Cole et al., 2013, 2014; Dodds et al., 2011; Power et al., 2011; Scolari et al., 2015). The activation of this fronto-parietal executive network (together with the insula and supplemental motor areas, also detected as part of this network) has been largely described as associated with voluntary emotion regulation (Ghomroudi et al., 2024; Messina et al., 2015; Monachesi et al., 2023; Morawetz et al., 2017). Thus, our findings suggest that high accepters could be more able in recruiting cognitive and attention regulatory brain networks to down-regulate emotion.
Among prefrontal areas, the dorsomedial prefrontal cortex (with extensions to the ACC and SMA) has deserved particular attention in the context of the research on mindfulness because it is involved in meditative acceptance-based processes (for a meta-analysis see Tomasino et al., 2013). Moreover, the DMPFC has been previously reported as more activated during emotion regulation tasks in individuals with higher trait mindfulness (Frewen et al., 2010; Modinos et al., 2010). Aside from emotion regulation, the DMPFC has been associated with evaluation of self-referential stimuli (Gusnard et al., 2001; Northoff & Bermpohl, 2004). For this reason, activity in DMPFC regions is thought to support meta-cognitive reflective awareness of internal states (Lutz et al., 2016; Olsson & Ochsner, 2008). Consistently, the higher gray matter concentration in high accepters observed in the present study may reflect the ability to be more aware of internal states.
The result of this study revealed a greater level of openness to experience among high accepters, as indicated by a statistically significant difference with a medium effect size. This suggests that high accepters tend to exhibit higher levels of curiosity, imagination, and open-mindedness, as well as a greater propensity to embrace unconventional ideas (Barrick et al., 2001). These individuals show a keen interest in both their inner thoughts and the external world (Costa & McCrae, 1992). The commonality observed between individuals with a high acceptance ability and openness to experience is their propensity for maintaining an open and curious mindset, continuously engaged with the flow of experiences (Banfi & Randall, 2022; Barańczuk, 2019; Giluk, 2009) High acceptance ability might play a role in facilitating and supporting the willingness to embrace new experiences associated with openness to experience, helping individuals maintain curiosity towards ongoing experiences. The results of this study show no significant relationship between high and low accepters and emotional intelligence, aligning with prior research findings (Mikolajczak et al., 2008; Zanella et al., 2022). However, this result contradicts our initial hypothesis, suggesting the necessity for in-depth further investigation and exploration.
Limitations and Future Directions
These findings offer valuable insights into the neural underpinnings of low and high accepters, which may have implications for the development of neuromodulation treatments targeting specific brain circuits to enhance acceptance abilities. However, there are some limitations to note. The reliance on self-report questionnaires in this study may have introduced biases due to participants' limited self-awareness. Additionally, the current study exclusively used structural data, and future research would benefit from integrating functional data to further explore these findings.
It was not possible to include demographic factors such as socioeconomic status, cultural background, and prior experience with mindfulness or meditation practices. While all participants were native German speakers, future research should include individuals with other ethnicities and cultural backgrounds and explore how these factors may moderate the relationship between acceptance and brain structure. We also acknowledge the imbalance in gender distribution, with a low number of female participants. Finally, the cross-sectional nature of the study restricts our ability to infer causality between neural structures and acceptance. Longitudinal studies tracking changes in brain structure over time in relation to acceptance are necessary to establish causal links and understand how acceptance may evolve with different interventions or experiences.
Acknowledgements
The authors would like to thank the Max Planck Institute for Human Cognitive and Brain Sciences (MPI CBS) in Leipzig for providing access to the LEMON dataset.
Declarations
Ethics Statement
The study was conducted in compliance with the Declaration of Helsinki. Ethical approval for the study was provided by the University of Leipzig's medical faculty ethics committee (reference number 154/13-ff).
Informed Consent Statement
Participants in the study provided written informed consent prior to data collection and agreed to the anonymous sharing of their data. No minors were involved in the study.
Competing Interests
The authors have no competing interests to declare.
Preregistration
This study is not preregistered.
Use of Artificial Intelligence Statement
AI was used for editing the manuscript to improve English language.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Met BSL Psychologie Totaal blijf je als professional steeds op de hoogte van de nieuwste ontwikkelingen binnen jouw vak. Met het online abonnement heb je toegang tot een groot aantal boeken, protocollen, vaktijdschriften en e-learnings op het gebied van psychologie en psychiatrie. Zo kun je op je gemak en wanneer het jou het beste uitkomt verdiepen in jouw vakgebied.
