Skip to main content
NCBI home page
European Journal of Psychotraumatology logoLink to European Journal of Psychotraumatology
. 2025 Sep 12;16(1):2551953. doi: 10.1080/20008066.2025.2551953

Brain network controllability in genetic risk, childhood abuse, and adult anxiety

Controlabilidad de las redes cerebrales en el riesgo genético, el abuso infantil y la ansiedad en la edad adulta

Tian Tian a, Min Chen b, Jicheng Fang a,CONTACT, Wenzhen Zhu a,✉
PMCID: PMC12434858  PMID: 40938193

ABSTRACT

Background: Network control theory can quantify controllability to evaluate how altered transitions between brain states contribute to cognitive, emotional, and behavioural challenges. Childhood abuse, influenced by genetics, is associated with disrupted network function, though the exact control processes are not yet understood.

Objective: This study aims to investigate the association between brain network controllability and childhood abuse experiences, and to elucidate its potential mediating role in the relationship among polygenic risk scores (PRS) for childhood abuse, childhood abuse experiences, and adult health outcomes.

Methods: This study measured the controllability of functional brain networks, including both average and modal controllability, in a cohort of 214 young adults with varied histories of childhood abuse. Participants also completed psychological assessments, whole-exome sequencing, and the calculation of PRS for childhood abuse. This study investigate the association between brain network controllability and childhood abuse. Furthermore, a mediation model was performed to explore the potential mediating role of brain network controllability in the relationship between genetic risk, childhood abuse experiences, and adult health outcomes.

Results: The controllability of the dorsal attention and sensorimotor networks, as well as the controllability of key ROIs within the sensorimotor, default mode, dorsal attention, visual, and control networks, demonstrated significant correlations with abuse scores. Despite no direct correlation between PRS and self-reported childhood abuse, indirect effects through the controllability of visual and control network regions were identified. The controllability of the left postcentral gyrus in the dorsal attention network mediated the relationship between childhood abuse and adult anxiety.

Conclusions: This study reveals that brain network controllability is a pivotal factor, not only bridging PRS and childhood abuse but also serving as a potential mediator between childhood trauma and adult anxiety, offering a new perspective on the neurobiology of childhood abuse-related psychopathology.

KEYWORDS: Childhood abuse, polygenic risk score, anxiety, controllability, functional networks, mediation analysis

HIGHLIGHTS

  1. Brain network controllability reveals the link between genetic risk for childhood abuse and actual experiences.

  2. The controllability of the left postcentral gyrus mediates the relationship between childhood abuse and adult anxiety.

  3. This study rovides new insights into the neurobiological mechanisms underlying childhood abuse-related mental health issues.

1. Introduction

Childhood abuse represents a critical social issue with profound and enduring consequences for mental and physical health (Chen, Yin et al., 2024; Hart et al., 2018; Kim et al., 2023). Accumulating evidence indicates that exposure to early-life adversity, such as physical, emotional, or sexual abuse, significantly elevates the lifetime risk of psychiatric disorders, including depression, schizophrenia, and post-traumatic stress disorder (Alkema et al., 2024; Florez et al., 2022; Kim et al., 2023). Recent advances in network neuroscience have refined the understanding of how childhood adversity shapes dynamic brain organisation, particularly through the lens of functional connectivity networks (the static statistical dependencies between regional neural time series) (Cao et al., 2024 Rakesh et al., 2023;) and network controllability (a distinct framework from control theory that quantifies the dynamic capacity of neural systems to transition between cognitive or affective states) (Stocker et al., 2023).

In past research, the plasticity of brain networks has been considered a potential mechanism by which childhood adversity affects adult mental health (Penza et al. (2003), Ibrahim et al. (2021). Brain networks are systems comprising multiple brain regions that work in concert during cognitive, emotional, and social tasks. In recent years, an increasing number of studies have shown that childhood abuse has a profound impact on brain network functioning. Childhood abuse has been observed to correlate with increased functional connectivity within the salience network (Rakesh et al., 2023). Furthermore, it has been implicated in the enhancement of connectivity within the attention network, as well as in the interplay between various resting-state networks (Cao et al., 2024). The functional connectivity of the language network with frontal, sensorimotor, and attentional networks has been shown to be influenced by maternal adverse experiences, interacting with sex and age, where girls exposed to mothers with a history of childhood abuse exhibited increased development of functional connectivity between the language network and visual networks, associated with social problems (Zhang et al., 2021). An earlier age of exposure to childhood abuse impacts the functional activation of cognitive control networks in adulthood (Seghete KL et al., 2018). While functional connectivity analyses have revealed altered coupling within brain networks following childhood abuse, such observations fall short of explaining how abused individuals develop impaired regulatory capacities. This gap is addressed by the concept of network controllability, a computational approach that measures the brain’s capacity to dynamically modulate activity within and across neural networks during task performance (Stocker et al., 2023). It quantifies the degree to which specific brain regions can influence the transitions of the entire brain’s functional state. Rooted in control theory and network neuroscience, controllability metrics like average controllability (propensity to drive transitions into easily reachable states) and modal controllability (ability to access difficult-to-reach states) (Chari et al., 2022) provide mechanistic insights into psychiatric vulnerabilities. Notably, altered controllability of the default mode, sensorimotor, attention, and frontoparietal networks has been implicated in depression and schizophrenia (Fang et al., 2022, 2023; Li et al., 2023), suggesting shared neural network substrates across stress-related disorders. Nevertheless, no prior work has examined whether childhood abuse disrupts brain network controllability to mediate psychiatric risk. Exploring this association could provide valuable insights into the complex interplay between environmental factors and neural processes in the development of psychopathology.

Self-reported childhood maltreatment experiences possess a significant genetic basis. Behavioural genetics has illuminated the heritability of reported childhood abuse, uncovering gene-environment interaction mechanisms (Dalvie et al., 2020). Research indicates that certain genetic variations can influence individual responses to adversity (Chen, Li et al., 2024; Li et al., 2024). This implies that an individual’s genetic makeup may influence their risk of experiencing childhood abuse. However, a single nucleotide polymorphism (SNP) or gene typically accounts for only a small fraction of the phenotypic variation, which is inadequate to capture the complexity of reported childhood abuse. Genome-wide association studies (GWAS) offer a promising approach for identifying individuals at risk for childhood abuse based on a polygenic inheritance model (Dalvie et al., 2020). The Polygenic Risk Score (PRS) is a widely used method in the follow-up analysis of GWAS data, calculated by summing the weighted effects of thousands of risk alleles in an individual’s genome to estimate the cumulative risk from many SNPs (Dudbridge, 2013). PRS can help identify those at higher risk of experiencing childhood abuse (Dalvie et al., 2020). Given that brain network function plays a crucial role in mediating the relationship between genetic factors and behavioural or experiential outcomes (Miller et al., 2018; Zhang et al., 2022), investigating how PRS influences brain network controllability, and consequently contributes to childhood abuse experiences, represents a compelling research direction. Specifically, PRS might reveal underlying genetic vulnerabilities and associated alterations in brain network controllability in individuals who are more susceptible to childhood abuse. This positions brain network function as a critical interface, potentially bridging innate genetic predispositions and subsequent childhood abuse experiences.

This study aims to investigate the association between specific indices of brain network controllability (average and modal controllability) within key functional networks, including the salience, default mode, control, sensorimotor, and attention networks, and self-reported childhood abuse experiences. Furthermore, it seeks to elucidate the potential mediating role of these specific network controllability measures in the relationship between PRS for childhood abuse and self-reported childhood abuse experiences. We also aim to explore whether these network controllability measures mediate the impact of childhood abuse on specific adult mental health outcomes, such as anxiety, depression, and personality. We hypothesise that brain network controllability serves as a pivotal factor: it may link genetic risk for adversity (PRS) to childhood abuse experiences and also mediate the impact of childhood trauma on poorer mental health outcomes. By examining these pathways, we hope to deepen our understanding of how genetics, brain function, and the environment shape individual experiences and vulnerability to psychopathology.

2. Material and methods

2.1. Study design and procedure

A flow diagram of the study design is succinctly shown in Figure 1. In this study, we recruited participants with varying degrees of childhood abuse experiences. Recruitment advertisements, outlining the study’s general purpose and participation requirements, were used across both online and offline channels. On one hand, advertisements were posted on social media platforms to attract potential participants. On the other hand, we collaborated with local universities to post recruitment posters on campus and recruit volunteers. After an initial telephone screening with the participants, appointments for MRI scanning were scheduled. Prior to the MRI scan, participants were scheduled on-site to complete the informed consent form, the screening information booklet, the MRI safety questionnaire, and the scales specific to this study, which were administered under the guidance of an attending physician with ten years of experience. Participants who met all recruitment criteria underwent the MRI scan. However, the scan was terminated if lesions or structural abnormalities were detected in the brain, or if the image quality remained consistently poor during the scanning process. Participants with qualified MRI images then had blood collected via venipuncture. Their samples were subsequently processed, aliquoted, and stored for future whole exome sequencing.

Figure 1.

Figure 1.

Open in a new tab

A flow diagram of the study design.

Note: PRS, polygenic risk score.

2.2. Recruitment criteria and questionnaires

To ensure the validity and reliability of our findings, particularly given the sensitivity of neuroimaging data and the nature of the psychological constructs assessed, we implemented rigorous inclusion criteria. The inclusion criteria for participants were as follows: 1) Ages between 18 and 30 years, Han ethnicity, right-handed, females were not pregnant and were not in their menstrual period on the day of scanning and behavioural assessments; 2) No contraindications for MRI and able to tolerate long-duration MRI scans, previous MRI scans did not reveal any intracranial abnormalities, no hearing impairments, colour blindness, or colour vision deficiencies; 3) No history of long-term smoking, excessive alcohol consumption, and no history of drug, substance abuse, or substance dependence; 4) No history of neuropsychiatric disorders: more than 5 minutes of consciousness loss (such as transient ischemic attacks, epilepsy, etc.), craniocerebral surgery, brain trauma (such as craniofacial fractures, intracranial hematomas, brain contusions, varying degrees of concussion), cerebrovascular diseases, encephalitis, major depressive disorder, schizophrenia, or other affective disorders; 5) No major physical illnesses (such as hypertension, hyperlipidemia, heart disease, diabetes, malignant tumours, kidney disease, autoimmune or hereditary metabolic diseases); 6) No family history of neuropsychiatric disorders; 7) No use of sedatives or non-prescription sleep aids in the last month, and currently not undergoing any medication treatment (including cold medicine, weight loss drugs, contraceptives); 8) No history of using antipsychotic drugs, antiepileptic drugs, psychostimulants, mood stabilisers, appetite suppressants, corticosteroids, antituberculosis drugs, or central antihypertensive medications; 9) Ensured adequate sleep the night before the experiment (at least 7 hours), prohibited the consumption of strong tea, caffeine, or alcoholic beverages on the day of the experiment, and ensured no intense physical activity; 10) No participants with familial relations were included in this study. These criteria were established with the aims of ensuring data quality, minimising potential confounding variables, preventing underlying pathologies or physiological states from influencing both the neuroimaging measures and the psychological assessments, and controlling for factors known to potentially affect brain function, genetics, behaviour, and mood. The human experiment was approved by the Ethical Committee of Tongji Hospital of Tongji Medical College of Huazhong University of Science and Technology. All subjects provided written informed consent in accordance with the Declaration of Helsinki.

For the childhood abuse phenotype, we utilised the Childhood Trauma Questionnaire (CTQ)-Short Form to assess the prevalence of emotional, physical, and sexual abuse experienced during childhood (Bernstein et al., 1997; Everaerd et al., 2016). Its Chinese version is a validated and reliable tool for measuring child maltreatment in Chinese populations (Wang et al., 2022). By providing quantitative data on childhood trauma, the CTQ aids researchers and clinicians in understanding and addressing trauma-related psychological and behavioural issues. The 25-item questionnaire uses a 5-point scale (1 = never true, 5 = very often true). An overall childhood abuse count score was derived by tallying items across the three abuse types. Higher scores indicate greater abuse severity. State and trait anxiety were measured using the Chinese version of the State-Trait Anxiety Inventory (STAI, Y form). This instrument comprises two parallel 20-item scales, each scored on a 4-point frequency scale. The STAI demonstrates strong psychometric properties and reliability within Chinese populations (Ma et al., 2013). Depressive symptoms over the past two weeks were assessed with the 21-item Chinese version of the Beck Depression Inventory-II (BDI-II). Items are rated on a 4-point severity scale, and this self-report measure shows good psychometric properties for evaluating depression in Chinese populations (Wang et al., 2020). Personality traits – novelty seeking, harm avoidance, and reward dependence – were evaluated using the Chinese adaptation of the Three-Dimensional Personality Questionnaire Version 4 (TPQ-4). This 100-item true/false questionnaire has been shown to possess good reliability in its Chinese version (Shao et al., 2020).

2.3. Genotyping and PRS estimation

Whole blood samples were extracted using the DNeasy Blood & Tissue Kit (QIAGEN). The library construction followed the Agilent exome capture sequencing protocol. Initially, the genomic DNA was fragmented, end-repaired, A-tailed at the 3’ end, and adapters were ligated. Subsequent steps included library amplification, hybridisation, post-hybridisation amplification, and finally, the construction of the sequencing library. To ensure library quality, the Qubit® 2.0 Fluorometer was used to measure the library concentration, and the Agilent 2100 was used to assess the library size. The constructed library was then sequenced using the Illumina HiSeq X platform with 2*150 bp reads. The BWA tool was used to align the reads from the fastq files to the human genome (hg19/GRCh37 version), followed by sorting with samtools and marking duplicates with picard. The GATK tool was then used for local realignment around insertions and deletions and base quality score recalibration. The resulting preprocessed bam files were analyzed using BEDtools and perl/python scripts to compile alignment information, including coverage and sequencing depth. The GATK Haplotype Caller was used to detect SNP and INDEL variants, and the GATK VQSR algorithm was applied to assess false positives, with credible variants marked as PASS. Copy number variations were analyzed using the eXome exome hidden Markov model. The ANNOVAR tool was used for variant annotation, including standard variant nomenclature, population database frequencies, disease database annotations, and functional mutation predictions. The PLINK genome-wide association study toolset was used for SNP data quality control. SNPs with a detection rate less than 90%, minor allele frequency less than 5%, and those not in Hardy – Weinberg equilibrium were excluded from the analysis.

We derived a PRS for self-reported childhood abuse from previously published GWAS as per (Dalvie et al., 2020), utilising the allele count methodology established by (Dudbridge, 2013). We computed an optimised PRS using the suggestive significance threshold (PT = 0.0354) as recommended by (Dalvie et al., 2020), which was derived from comprehensive data analysis for predicting childhood abuse and demonstrated considerable practical utility. This optimised PRS encapsulates the cumulative impact of SNPs that are most indicative of self-reported childhood abuse.

2.4. MRI acquisition and data preprocessing procedures

All scanning procedures were conducted on a 3.0-Tesla MRI system (Discovery MR750, General Electric). To minimise head movement, a snug and comfortable foam pad was employed, and earplugs were provided to reduce scanner noise. Functional images were acquired via an echo planar imaging sequence using these parameters: repetition time (TR)/echo time (TE) of 2000/30 ms, a field of view (FOV) of 220 mm × 220, 3 mm slice thickness, 90° flip angle, a 64 × 64 matrix, no inter-slice gaps, resulting voxel dimensions of 3.43 × 3.43 × 3.43 mm³, 36 interleaved transverse slices, and a total of 185 volumes. Complementing this, high-resolution 3D T1-weighted brain volume images were also collected with the following settings: TR/TE of 8.16/3.18 ms, inversion time (TI) of 450 ms, 12° flip angle, FOV of 256 mm × 256 mm, 256 × 256 matrix, 1 mm slice thickness, contiguous slices yielding 1 × 1 × 1 mm³ voxels, and 188 sagittal slices.

For preprocessing of the functional data, the initial ten volumes for each participant were discarded. The remaining 175 volumes underwent slice time correction. Head motion was assessed to correct for any participant movement. To standardise the functional images, individual structural scans were first linearly co-registered to the mean functional image obtained after motion correction. These structural images were then segmented and nonlinearly registered to the Montreal Neurological Institute (MNI) template. Using the parameters derived from this nonlinear registration, the motion-corrected functional volumes were subsequently transformed into MNI space. The functional data were resampled to achieve uniform 3 mm³ voxels. Finally, a bandpass filter (0.01–0.08 Hz) was applied. Additionally, regression analysis was used to remove the influence of several confounds, including 24 motion parameters as well as the average BOLD signals from the ventricles, white matter, and the global brain signal.

Time series were derived from the pre-processed functional MRI images within defined anatomic regions (Schaefer et al., 2018). The current investigation encompassed 100 regions of interest (ROIs) spanning seven distinct brain networks: the limbic network, default mode network, dorsal attention network, salience network, sensorimotor network, control network, and visual network (Figure 2). The specific labels and abbreviations for these ROIs can be found in Supplementary Table 1. The most commonly used measures of network controllability are modal controllability and average controllability. Average controllability is a measure of a node’s ability to steer the brain into all easily reachable states, taking into account the average input energy cost. Average controllability is analogous to the trace of the Gramian matrix and is inversely related to the control energy needed to induce transitions in brain states. Brain regions with high average controllability can facilitate transitions in functional states with less input energy, suggesting that they can more readily shift into desired network configurations. Modal controllability quantifies the ease with which a single control node can drive the brain into hard-to-reach states. Regions with high modal controllability can more easily push the dynamics of the brain network into less accessible states, imposing higher energy costs for accomplishing complex, target-specific tasks. For the details on the derivation of these metrics, please refer to the literature (Li et al., 2023).

Figure 2.

Figure 2.

Open in a new tab

Distribution map of 100 ROIs across seven functional brain networks.

Note: ROIs, regions of interest.

2.5. Statistical analysis

We utilised G*Power (3.1.9.7), a user-friendly and free tool, to conduct a priori power analyses to ensure our study had enough statistical power. Our conservative approach involved expecting a medium effect size (f² = 0.15), seeking a power of 0.8, and using a significance level of 0.05.

Statistical analyses of the demographic data were conducted using the Statistical Package for the Social Sciences, version 19.0. To assess the normal distribution of both behavioural and imaging data, the Kolmogorov–Smirnov test was applied. The reliability of the scales was evaluated using Cronbach’s alpha coefficient. After controlling for sex, age, years of education, and head motion, we employed Pearson correlation analysis to examine the relationship between changes in the controllability of each network and ROI within the networks and the degree of childhood abuse. To account for multiple comparisons, a false discovery rate (FDR) correction using the Benjamini-Hochberg procedure was applied to the correlation analysis. Significantly associated networks or nodes were further investigated in subsequent mediation analyses.

In this study, we utilised a three-variable mediation model via the SPSS macro (available at http://www.processmacro.org/index.html) to elucidate the interplay between PRS, brain controllability, and abuse scores. This approach may enhance our comprehension of the link between inherited risk and self-reported childhood abuse, even when no direct association is apparent. Additionally, we employed a mediation analysis model to examine the relationship between childhood abuse scores, brain controllability, and behavioural assessments, aiming to discern the role of brain network controllability in the impact of childhood abuse on adult anxiety, depression, and personality traits. The bootstrapping technique was used to evaluate the significance of the mediating effects.

3. Results

3.1. Demographic and behavioural characteristics

Demographic details and behavioural attributes of the 214 participants are encapsulated in Table 1. Of these, 175 participants underwent peripheral blood collection followed by whole exome sequencing. The mean PRS for self-reported childhood abuse was 0.37, with a standard deviation of 0.26, spanning a range from −0.30 to 1.09. The Cronbach’s alpha coefficient for the entire scale was acceptably high (α = 0.71). Both imaging characteristics and behavioural scores conformed to a normal distribution. No significant gender disparities were observed in PRS, abuse scores, age, educational attainment, or behavioural scores.

Table 1.

Demographic and behavioural characteristics of 214 subjects.

Demographics and behaviors Mean (Standard deviation) Range
Gender (female/male) 157/57 NA
Age (years) 24.05 (1.90) 20–30
Education level (years) 17.65 (1.54) 13–22
Emotional abuse 6.16 (1.48) 5–11
Sexual abuse 5.20 (0.60) 5–9
Physical abuse 5.51 (1.28) 5–14
Overall childhood abuse 16.87 (2.53) 15–30
BDI 5.37 (4.27) 0–22
State anxiety 32.66 (8.08) 20–57
Trait anxiety 36.57 (7.70) 21–63
NS 13.88 (4.41) 4–29
HA 15.25 (5.66) 2–30
RD 19.11 (3.36) 10–27
Open in a new tab

Note: BDI, Beck Depression Inventory; HA, Harm Avoidance; NA, not applicable; NS, Novelty Seeking; RD, Reward Dependence; SD, standard deviation.

3.2. The impact of childhood abuse on the network controllability and the controllability of brain regions within functional networks

Overall, the average controllability of the dorsal attention network was found to show a significant positive correlation with childhood physical abuse (r = 0.225, PFDR = 0.01), while the average controllability of the sensorimotor network demonstrated a significant negative correlation with childhood physical abuse (r = −0.217, PFDR = 0.01).

We found that the controllability of specific ROIs within the sensorimotor network, default mode network, dorsal attention network, visual network, and control network was significantly associated with childhood abuse scores (Figure 3). Negative correlations were identified between the average controllability of ROIs within the sensorimotor network and the physical abuse score (r = −0.203, PFDR = 0.02; r = −0.201, PFDR = 0.02). Similarly, the average controllability of an ROI in the left prefrontal cortex, part of the default mode network, was inversely associated with the physical abuse score (r = −0.195, PFDR = 0.04). The average controllability of the ROI located in the left precentral ventral gyrus, labelled LH_PrCv_1, and the ROI in the right postcentral gyrus, labelled RH_Post_3, within the dorsal attention network, respectively displayed a significant positive correlation with the degree of childhood physical abuse (r = 0.238, PFDR = 0.01; r = 0.213, PFDR = 0.02). Moreover, the average controllability of ROI labelled LH_Post_2, situated in the left postcentral gyrus of the dorsal attention network, revealed a positive correlation with both childhood physical abuse (r = 0.197, PFDR = 0.03) and total abuse scores (r = 0.201, PFDR = 0.02), while its modal controllability was negatively linked to physical (r = −0.213, PFDR = 0.02) and total (r = −0.194, PFDR = 0.03) abuse scores. Additionally, the average controllability of ROI labelled LH_Post_4, in the left postcentral gyrus of the dorsal attention network, showed a positive correlation with scores of childhood emotional abuse (r = 0.189, PFDR = 0.04) and total abuse (r = 0.200, PFDR = 0.02). The average controllability of ROI located in the visual network, labelled LH_1, displayed a significant negative correlation with the degree of childhood physical abuse (r = −0.150, PFDR = 0.04). The average controllability of an ROI in the left parietal cortex, part of the control network, was positively associated with the physical abuse score (r = 0.163, PFDR = 0.03).

Figure 3.

Figure 3.

Open in a new tab

Brain network ROIs associated with childhood abuse. The controllability of ROIs within the sensorimotor, default mode, dorsal attention, visual, and control networks exhibited significant correlations with childhood abuse scores.

Note: LH, left hemisphere; Par, parietal; PFC, prefrontal cortex; Post, postcentral; PrCv, precentral ventral; RH, right hemisphere; ROI, region of interest.

The fifteen significant correlations between various types of childhood abuse and network/ROI controllability, as summarised in Table 2, provide the basis for subsequent mediation analyses examining the role of controllability in the relationship among PRS for childhood abuse, childhood abuse experiences, and adult health outcomes.

Table 2.

Correlations between childhood abuse types and network controllability across different brain regions.

Network ROI Controllability Type Abuse Type Correlation (r) PFDR
Dorsal Attention Overall Mean Average Physical 0.225 0.01
Sensorimotor Overall Mean Average Physical −0.217 0.01
Sensorimotor LH_6 Average Physical −0.203 0.02
Sensorimotor RH_5 Average Physical −0.201 0.02
Default Mode LH_PFC_7 Average Physical −0.195 0.04
Dorsal Attention LH_PrCv_1 Average Physical 0.238 0.01
Dorsal Attention RH_Post_3 Average Physical 0.213 0.02
Dorsal Attention LH_Post_2 Average Physical 0.197 0.03
Dorsal Attention LH_Post_2 Average Total 0.201 0.02
Dorsal Attention LH_Post_2 Modal Physical −0.213 0.02
Dorsal Attention LH_Post_2 Modal Total −0.194 0.03
Dorsal Attention LH_Post_4 Average Emotional 0.189 0.04
Dorsal Attention LH_Post_4 Average Total 0.200 0.02
Visual LH_1 Average Physical −0.150 0.04
Control LH_Par_1 Average Physical 0.163 0.03
Open in a new tab

Note: FDR, LH, left hemisphere; Par, parietal; PFC, prefrontal cortex; Post, postcentral; PrCv, precentral ventral; RH, right hemisphere; ROI, region of interest.

3.3. The role of brain network controllability in the impact of PRS on childhood abuse

The subsequent mediation analyses examined the role of 15 network/ROI controllability as a mediator, linking the independent variable (PRS related to childhood abuse risk) to childhood abuse. G*Power software indicated that aminimum sample size of 143 would be required. Our actual sample size (N = 175) exceeded this threshold, thereby providing adequate power for the planned analyses.

The mediation effect analysis revealed that although there were no significant direct effects between the PRS and physical abuse, significant indirect effects were observed, mediated by the average controllability of visual (Figure 4a) and control (Figure 4b) network regions. We found that the average controllability of ROI located in the visual network, labelled LH_1, was a significant mediator (Index = −0.032, 95% CI = −0.075 to –0.011) between PRS and physical abuse scores (as shown in Figure 4a). The average controllability of ROI located in the parietal cortex of control network, labelled LH_Par_1, significantly mediated the relationship (Index = −0.029, 95% CI = −0.124 to –0.001) between PRS and physical abuse scores (as shown in Figure 4b). The model indices and corresponding regression coefficients are presented in Supplementary Tables 2 and 3. No significant mediating role of network or intra-network ROI controllability was detected in the relationship between PRS and other subtypes of childhood abuse, nor in the relationship with the overall abuse index.

Figure 4.

Figure 4.

Open in a new tab

The role of brain network controllability in the impact of PRS on physical abuse. The PRS influence on physical abuse was mainly mediated through ROIs within the visual network and the control network. There was a significant positive effect from the PRS to average controllability of the visual network region, a significant negative effect from the average controllability of the visual network region to physical abuse, and a significant indirect effect rather than direct effect from the PRS to physical abuse (Part a). There was a significant negative effect from the PRS to average controllability of the control network region, a significant positive effect from the average controllability of the control network region to physical abuse, and a significant indirect effect rather than direct effect from the PRS to physical abuse (Part b).

Note: LH, left hemisphere; Par, parietal; PRS, polygenic risk score; ROI, region of interest.

3.4. The role of brain network controllability in the impact of childhood abuse on its consequences in adulthood

The subsequent mediation analyses examined the role of 15 network/ROI controllability as a mediator, linking the independent variables (childhood abuse types) to adult mental healthy. G*Power software indicated that a minimum sample size of 153 would be required. Our actual sample size (N = 214) exceeded this threshold, thereby providing adequate power for the planned analyses.

The average controllability of ROI LH_Post_4, located in the left postcentral gyrus of the dorsal attention network, played a pivotal mediating role in the relationship between childhood abuse and both adult state anxiety (Figure 5a) and trait anxiety (Figure 5b). This ROI’s controllability significantly mediated multiple pathways: the link between emotional abuse and state anxiety (Index = 0.049, 95% CI = 0.007–0.118; model indices and regression coefficients in Supplementary Table 4), the relationship between overall abuse and state anxiety (Index = 0.052, 95% CI = 0.006–0.120; model indices and regression coefficients in Supplementary Table 5), the association between emotional abuse and trait anxiety (Index = 0.036, 95% CI = 0.005–0.095; model indices and regression coefficients in Supplementary Table 6), and the connection between overall abuse and trait anxiety (Index = 0.040, 95% CI = 0.004–0.107; model indices and regression coefficients in Supplementary Table 7). No significant mediating role of network or intra-network ROI controllability was detected in the relationship between childhood abuse and depression/personality traits.

Figure 5.

Figure 5.

Open in a new tab

The role of brain network controllability in the impact of childhood abuse on anxiety. (Part a) The left postcentral gyrus within the dorsal attention network mediates the relationship between childhood abuse and adult state anxiety. (Part b) The left postcentral gyrus within the dorsal attention network mediates the relationship between childhood abuse and adult trait anxiety. The figure depicts the pathway by which childhood abuse impacts adult anxiety through alterations in brain network controllability. There were significant positive effects from the abuse scores to the average controllability of ROI in the left postcentral gyrus of the dorsal attention network, significant positive effects from the average controllability of ROI in the left postcentral gyrus to the anxiety scores, and significant total (c), direct (c’), and indirect (ab) effects from the abuse scores to the anxiety scores.

Note: LH, left hemisphere; Post, postcentral; ROI, region of interest.

4. Discussion

This study found a significant positive correlation between the average controllability of the dorsal attention network and childhood physical abuse, whereas a significant negative correlation was observed between the average controllability of the sensorimotor network and childhood physical abuse. The controllability of ROIs within the sensorimotor network, default mode network, dorsal attention network, visual network, and control network was significantly associated with childhood abuse scores. Although there was no direct correlation between polygenic risk and self-reported childhood abuse, it indirectly affected the abuse experience through the controllability of visual and control network regions. The findings emphasise the complexity of the relationship between genetic risk and childhood abuse and point to the potential role of brain network controllability in this relationship, providing a new perspective for understanding the neurobiological basis of childhood abuse. Furthermore, the average controllability of the ROI in the left postcentral gyrus of the dorsal attention network was not only positively correlated with childhood emotional abuse and total abuse scores but also significantly mediated the impact of childhood abuse experiences on adult state anxiety and trait anxiety. These findings underscore the role of brain network controllability in the association between childhood abuse and adult anxiety, offering a fresh theoretical perspective on the impact of early adversity on brain function and mental well-being.

In previous research, a wealth of literature has focused on the long-term impacts of childhood abuse on brain structure and functional connectivity. These studies have consistently found that childhood abuse is associated with structural abnormalities and alterations in functional connectivity across multiple brain regions (Gold et al., 2016; Paquola et al., 2017). Departing from the traditional approach, this study adopts a novel perspective of network controllability, revealing a deeper level of impact of childhood physical abuse on brain functional networks. The dorsal attention network is primarily responsible for processing and maintaining information from the external environment, as well as tasks involving spatial attention and visual search. The main function of this network is to monitor the external environment for the timely detection and processing of potential stimuli or threats. The study found that individuals who experienced more physical abuse during childhood exhibited significantly increased average controllability in the dorsal attention network, and they also reported higher levels of anxiety in adulthood. This may be due to the constant need for these individuals to adjust their attention in response to potential threats in a chronically unsafe environment. The increased average controllability likely signifies an enhanced ability to quickly modulate attention in the face of potential threats, accompanied by a heightened state of alertness. This interpretation is supported by evidence that early abuse is associated with greater connectivity in the dorsal attention network, suggesting that adults who have experienced childhood abuse may also have a preferential sensitivity to environmental cues (Korgaonkar et al., 2023). The literature also supports this view, indicating that heightened environmental vigilance due to exposure to high levels of threat in early life can lead to attentional biases and hypervigilance to threat, which is typically considered indicative of anxiety disorders (Mobini & Grant, 2007). The sensorimotor network is primarily responsible for processing sensory information and executing motor tasks. This study found a negative correlation between childhood physical abuse and the average controllability of the sensorimotor network. This suggests that individuals who suffered physical abuse during childhood may have deficiencies in the regulation of neurophysiological states related to the sensorimotor network. The literature also mentions that physical abuse leads to abnormal connectivity in the sensorimotor cortex (Diez et al., 2021), affecting behavioural and emotional regulation. By examining network controllability, this study reveals the distinct effects of childhood physical abuse on the dorsal attention and sensorimotor networks. This finding provides a new perspective on understanding the complexity of the impact of childhood abuse on brain functional networks and offers a scientific basis for the development of more effective intervention strategies in the future.

This study not only explored the controllability of the overall network but also delved into specific ROI levels, primarily discovering a significant correlation between the controllability of the sensorimotor and prefrontal cortices and different types of childhood abuse scores. This nuanced analysis helps to more accurately understand the differential impact of childhood abuse experiences on the functioning of various brain regions within networks. Although the precentral gyrus and postcentral gyrus, responsible for processing sensory input and generating motor output, are not core components of the dorsal attention network, they are functionally closely connected to it, jointly participating in the regulation of attention and action. The study revealed that childhood abuse experiences were positively correlated with the average controllability of the sensorimotor cortex within the high-level dorsal attention network related to motor control and execution, but negatively correlated with brain regions in the sensorimotor network responsible for processing basic sensory information and executing motor commands. This may uncover a functionally system-dependent effect of childhood abuse on network controllability, indicating that childhood abuse experiences have a profoundly complex impact on brain function. The influence of childhood abuse experiences on brain regions within the processing system, which includes the visual, sensorimotor, and default mode networks, may be quite the opposite to that on the control system, such as the fronto-parietal/control and attention networks (Power et al., 2011). This is an interesting topic worthy of further exploration in the future.

In this study, we unveiled the complex relationship between PRS and childhood abuse. Although statistical analysis did not find a direct connection between PRS and childhood abuse, the mediation effect analysis revealed that genetic variations might indirectly relate to the experience of childhood abuse by affecting the average controllability of brain network nodes. This finding highlights that the role of genetic factors in the risk of individual childhood abuse is not direct and singular but potentially connected to childhood abuse experiences through the regulation of neural plasticity in the brain. Notably, the participants in this study were recruited from the community, and their scores of childhood abuse experiences were generally low, with no participants having severe mental illnesses such as PTSD or major depression. Moderate brain network plasticity is crucial for individuals to adapt to environmental changes and regulate behaviour (Kolb & Gibb, 2011; Nava & Roder, 2011). We found that individuals with a high genetic risk of abuse may have undergone moderate remodelling in the visual network areas, which enhanced their ability to recognise potential danger signals and might have reduced the risk of experiencing childhood abuse. Previous research has shown that visual processing areas can be remodelled to better adapt to danger signals in the environment (Meaux et al., 2019), further confirming this point. The control network, involving the prefrontal cortex and parietal regions, primarily responsible for executive functions, decision-making, and inhibiting inappropriate behaviour, has shown that moderate plasticity in these areas is extremely important for coping with stress and adversity (Hermans et al., 2024; Sinha et al., 2016). For individuals with high PRS, moderate remodelling in the control network areas may help improve impulse control and emotional regulation, both of which are key factors in preventing childhood abuse, potentially reducing the risk of individuals suffering childhood abuse. In summary, our research reveals the indirect link between genetic risk and childhood abuse, emphasising the significant role of brain network plasticity in modulating this relationship. Network functional plasticity might act as a protective factor, mitigating the negative impact of genetic risk, depending on individual differences and other moderating variables. These findings not only deepen our understanding of the genetic-brain-environment pathway but also provide new perspectives and strategies for the prevention and intervention of childhood abuse. Future research should further explore these mechanisms to more effectively protect children from the harm of abuse.

Additionally, the study found that the ROI in the left postcentral gyrus of the dorsal attention network not only positively correlated with childhood emotional abuse and total abuse scores but also mediated the impact of these experiences on adult anxiety behaviour. The controllability of the left postcentral gyrus in the dorsal attention network may serve as a biomarker reflecting an individual’s ability to cope with stress and trauma from childhood abuse. The postcentral gyrus plays a crucial role in processing somatosensory information, affecting an individual’s ability to handle anxiety-related bodily sensations (Clauss et al., 2019). Studies have shown that this brain region is closely associated with enhanced sensory sensitivity and persistent hypervigilance in individuals who have experienced childhood trauma (Tymofiyeva et al., 2022). Hypervigilance, a core feature of anxiety disorders, reflects an individual’s ongoing concern with potential threats (Clauss et al., 2019). Therefore, increased controllability of the postcentral gyrus may exacerbate an individual’s sensitivity to and response patterns towards environmental threats, potentially leading to or worsening anxiety behaviour. Moreover, the literature reports that hypervigilance to threat due to attentional biases is typically considered indicative of anxiety disorders (Mobini & Grant, 2007), which also supports our findings. This finding is of great significance for understanding the long-term impact of childhood abuse on mental health. It suggests that early adverse experiences may have long-lasting effects on an individual’s mental health through alterations in brain function controllability. This provides a new perspective on understanding the neurobiological basis of anxiety disorders and may offer new targets for future intervention strategies. However, alternative interpretations must also be considered. For example, reverse causality remains a possibility: pre-existing or concurrent anxiety traits could have influenced the measured network controllability, independent of abuse experiences. Moreover, the observed relationships might be influenced by confounding factors, such as current adult stressors, coping mechanisms, or socioeconomic status, which could independently affect both abuse reporting and brain network properties. Further research is clearly needed to disentangle these complexities.

This study has several limitations. Firstly, it is important to note that our sample included individuals with relatively mild abuse histories and those with no childhood abuse experience. Specifically, no participant scored in the severe range on the CTQ subscales for emotional, physical, or sexual abuse. This characteristic of our sample means that while our findings are informative for understanding the impact within this specific, less severe range, they may not fully capture the effects observed in individuals with more severe abuse experiences, which could manifest differently in terms of magnitude or nature. Consequently, further studies including participants with a broader range of trauma severity, particularly more severe cases, are warranted to provide a more comprehensive picture. Secondly, a significant limitation is the retrospective nature of the childhood abuse reporting. Childhood abuse was assessed using self-report measures, which inherently carries the potential for recall bias. Participants may have difficulty accurately remembering past events, or their recall could be influenced by current mental health status or other factors, potentially affecting the accuracy of the reported abuse experiences. Thirdly, the current study’s cross-sectional design is a limitation, as it precludes establishing definitive causal pathways. Future research incorporating longitudinal data would be needed to more deeply understand the dynamic interplay between brain network controllability, childhood trauma, and adult anxiety. Lastly, this study did not explore the impact of resilience on the relationship between childhood trauma and adult mental health, a factor that may play a significant moderating role in this complex relationship. Resilience has a notable influence on individuals’ability to respond to and recover from trauma; therefore, investigating how resilience interacts with brain network controllability and affects the connection between childhood trauma and adult mental health is crucial for a comprehensive understanding of this relationship.

5. Conclusions

Overall, this study not only uncovers the potential mediating role of brain network controllability in the association between genetic risk for adversity and childhood abuse experiences but also highlights its critical role in the relationship between childhood abuse and adult anxiety. These findings provide valuable insights into the neurobiological underpinnings of childhood abuse, which may inform future research and potential therapeutic interventions aimed at mitigating the lasting effects of early trauma on mental well-being.

Supplementary Material

supplementary materials.docx
ZEPT_A_2551953_SM1770.docx (43.9KB, docx)

Acknowledgments

The authors would like to express sincere gratitude to Dr. Ning Zheng at Clinical and Technical Support, Philips Healthcare, Wuhan, China, for useful discussions.

Funding Statement

This study was supported by the National Natural Science Foundation of China (No. 82471965), the Regional Innovation and Development Joint Fund of National Natural Science Foundation of China (No. U22A20354), and National Key Research and Development Program of China (No. 2022YFC2406903). The funders of the study had no role in study design, planning, data analysis, data interpretation, or writing of the report.

Author contributions

Conception and design: Jicheng Fang, Wenzhen Zhu, and Tian Tian; Administrative support: Wenzhen Zhu and Tian Tian; Provision of study materials or patients: Tian Tian and Min Chen; Collection and assembly of data: Min Chen, Jicheng Fang, and Tian Tian; Data analysis and interpretation: Jicheng Fang and Tian Tian; Manuscript writing: All authors.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Ethics approval and consent to participate

The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study involving human participants was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by ethics board of Tongji Hospital of Tongji Medical College of Huazhong University of Science and Technology. Informed consent was documented in writing from the participants involved.

Data availability statement

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Supplemental Material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/20008066.2025.2551953.

References

  1. Alkema, A., Marchi, M., van der Zaag, J. A. J., van der Sluis, D., Warrier, V., Genetic, R., Ophoff, R. A., Kahn, R. S., Cahn, W., Hovens, J. G. F. M., Riese, H., Scheepers, F., Penninx, B. W. J. H., Cecil, C., Oldehinkel, A. J., Vinkers, C. H., & Boks, M. P. M. (2024). Childhood abuse v. Neglect and risk for major psychiatric disorders. Psychological Medicine, 54(8), 1598–1609. 10.1017/S0033291723003471 [DOI] [PubMed] [Google Scholar]
  2. Bernstein, D. P., Ahluvalia, T., Pogge, D., & Handelsman, L. (1997). Validity of the Childhood Trauma Questionnaire in an adolescent psychiatric population. Journal of the American Academy of Child & Adolescent Psychiatry, 36(3), 340–348. 10.1097/00004583-199703000-00012 [DOI] [PubMed] [Google Scholar]
  3. Cao, X., Wang, Z., Chen, Y., Consortium the, N., & Zhu, J. (2024). Childhood maltreatment and resting-state network connectivity: The risk-buffering role of positive parenting. Development and Psychopathology, 1–12. [DOI] [PubMed] [Google Scholar]
  4. Chari, A., Seunarine, K. K., He, X., Tisdall, M. M., Clark, C. A., Bassett, D. S., Scott, R. C., & Rosch, R. E. (2022). Drug-resistant focal epilepsy in children is associated with increased modal controllability of the whole brain and epileptogenic regions. Communications Biology, 5(1), 394. 10.1038/s42003-022-03342-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Chen, P., Li, Y., Zadrozny, S., Seifer, R., & Belger, A. (2024). Polygenic risk, childhood abuse and gene x environment interactions with depression development from middle to late adulthood: A U. S. National life-course study. Preventive Medicine, 185, 108048. [DOI] [PubMed] [Google Scholar]
  6. Chen, S., Yin, Y., Zhang, Y., Jiang, W., Hou, Z., & Yuan, Y. (2024). Childhood abuse influences clinical features of major depressive disorder by modulating the functional network of the right amygdala subregions. Asian Journal of Psychiatry, 93, 103946. 10.1016/j.ajp.2024.103946 [DOI] [PubMed] [Google Scholar]
  7. Clauss, J. A., Avery, S. N., Benningfield, M. M., & Blackford, J. U. (2019). Social anxiety is associated with BNST response to unpredictability. Depression and Anxiety, 36(8), 666–675. 10.1002/da.22891 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dalvie, S., Maihofer, A. X., Coleman, J. R. I., Bradley, B., Breen, G., Brick, L. A., Chen, C.-Y., Choi, K. W., Duncan, L. E., Guffanti, G., Haas, M., Harnal, S., Liberzon, I., Nugent, N. R., Provost, A. C., Ressler, K. J., Torres, K., Amstadter, A. B., Bryn Austin, S., … Orcutt, H. K. (2020). Genomic influences on self-reported childhood maltreatment. Translational Psychiatry, 10(1), 38. 10.1038/s41398-020-0706-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Diez, I., Larson, A. G., Nakhate, V., Dunn, E. C., Fricchione, G. L., Nicholson, T. R., Sepulcre, J., & Perez, D. L. (2021). Early-life trauma endophenotypes and brain circuit-gene expression relationships in functional neurological (conversion) disorder. Molecular Psychiatry, 26(8), 3817–3828. 10.1038/s41380-020-0665-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Dudbridge, F. (2013). Power and predictive accuracy of polygenic risk scores. PLoS Genetics, 9(3), e1003348. 10.1371/journal.pgen.1003348 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Everaerd, D., Klumpers, F., Zwiers, M., Guadalupe, T., Franke, B., van Oostrom, I., Schene, A., Fernández, G., & Tendolkar, I. (2016). Childhood abuse and deprivation are associated with distinct sex-dependent differences in brain morphology. Neuropsychopharmacology, 41(7), 1716–1723. 10.1038/npp.2015.344 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fang, F., Godlewska, B., Cho, R. Y., Savitz, S. I., Selvaraj, S., & Zhang, Y. (2022). Effects of escitalopram therapy on functional brain controllability in major depressive disorder. Journal of Affective Disorders, 310, 68–74. 10.1016/j.jad.2022.04.123 [DOI] [PubMed] [Google Scholar]
  13. Fang, F., Godlewska, B., Selvaraj, S., & Zhang, Y. (2023). Predicting antidepressant treatment response using functional brain controllability analysis. Brain Connectivity, 13(2), 107–116. 10.1089/brain.2022.0027 [DOI] [PubMed] [Google Scholar]
  14. Florez, I. A., Mekawi, Y., Hunnicutt-Ferguson, K., Visser, K. F., Clunie, A. M., Dunn, S. E., & Kaslow, N. J. (2022). Childhood abuse, posttraumatic stress symptoms, and alcohol misuse among African-American women. Journal of Ethnicity in Substance Abuse, 21(1), 174–196. 10.1080/15332640.2020.1725707 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Gold, A. L., Sheridan, M. A., Peverill, M., Busso, D. S., Lambert, H. K., Alves, S., Pine, D. S., & McLaughlin, K. A. (2016). Childhood abuse and reduced cortical thickness in brain regions involved in emotional processing. Journal of Child Psychology and Psychiatry, 57(10), 1154–1164. 10.1111/jcpp.12630 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hart, H., Lim, L., Mehta, M. A., Curtis, C., Xu, X., Breen, G., Simmons, A., Mirza, K., & Rubia, K. (2018). Altered functional connectivity of fronto-cingulo-striatal circuits during error monitoring in adolescents with a history of childhood abuse. Frontiers in Human Neuroscience, 12, 7. 10.3389/fnhum.2018.00007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hermans, E. J., Hendler, T., & Kalisch, R. (2024). Building resilience: The stress response as a driving force for neuroplasticity and adaptation. Biological Psychiatry, 97(4), 330–338. 10.1016/j.biopsych.2024.10.016. [DOI] [PubMed] [Google Scholar]
  18. Ibrahim, P., Almeida, D., Nagy, C., & Turecki, G. (2021). Molecular impacts of childhood abuse on the human brain. Neurobiology of Stress, 15, 100343. 10.1016/j.ynstr.2021.100343 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kim, S. Y., An, S. J., Han, J. H., Kang, Y., Bae, E. B., Tae, W. S., & Ham, B.-J. (2023). Childhood abuse and cortical gray matter volume in patients with major depressive disorder. Psychiatry Research, 319, 114990. 10.1016/j.psychres.2022.114990 [DOI] [PubMed] [Google Scholar]
  20. Kolb, B., & Gibb, R. (2011). Brain plasticity and behaviour in the developing brain. Journal of the Canadian Academy of Child and Adolescent Psychiatry, 20(4), 265–276. [PMC free article] [PubMed] [Google Scholar]
  21. Korgaonkar, M. S., Breukelaar, I. A., Felmingham, K., Williams, L. M., & Bryant, R. A. (2023). Association of neural connectome with early experiences of abuse in adults. JAMA Network Open, 6(1), e2253082. 10.1001/jamanetworkopen.2022.53082 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Li, Q., Yao, L., You, W., Liu, J., Deng, S., Li, B., Luo, L., Zhao, Y., Wang, Y., Wang, Y., Zhang, Q., Long, F., Sweeney, J. A., Gu, S., Li, F., & Gong, Q. (2023). Controllability of functional brain networks and its clinical significance in first-episode schizophrenia. Schizophrenia Bulletin, 49(3), 659–668. 10.1093/schbul/sbac177 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Li, Y., Jia, H., Wei, X., Zhen, S., He, S. C., & Zhang, X. Y. (2024). Interaction of childhood trauma with BDNF and FKBP5 gene polymorphisms in predicting burnout in general occupational groups. European Archives of Psychiatry and Clinical Neuroscience, Online ahead of print. 10.1007/s00406-024-01825-9. [DOI] [PubMed] [Google Scholar]
  24. Ma, W. F., Liu, Y. C., Chen, Y. F., Lane, H. Y., Lai, T. J., & Huang, L. C. (2013). Evaluation of psychometric properties of the Chinese mandarin version state-trait anxiety inventory Y form in Taiwanese outpatients with anxiety disorders. Journal of Psychiatric and Mental Health Nursing, 20(6), 499–507. 10.1111/j.1365-2850.2012.01945.x [DOI] [PubMed] [Google Scholar]
  25. Meaux, E., Sterpenich, V., & Vuilleumier, P. (2019). Emotional learning promotes perceptual predictions by remodeling stimulus representation in visual cortex. Scientific Reports, 9(1), 16867. 10.1038/s41598-019-52615-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Miller, J. A., Scult, M. A., Conley, E. D., Chen, Q., Weinberger, D. R., & Hariri, A. R. (2018). Effects of schizophrenia polygenic risk scores on brain activity and performance during working memory subprocesses in healthy young adults. Schizophrenia Bulletin, 44(4), 844–853. 10.1093/schbul/sbx140 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Mobini, S., & Grant, A. (2007). Clinical implications of attentional bias in anxiety disorders: An integrative literature review. Psychotherapy: Theory, Research, Practice, Training, 44(4), 450–462. 10.1037/0033-3204.44.4.450 [DOI] [PubMed] [Google Scholar]
  28. Nava, E., & Roder, B. (2011). Adaptation and maladaptation insights from brain plasticity. Progress in Brain Research, 191, 177–194. 10.1016/B978-0-444-53752-2.00005-9 [DOI] [PubMed] [Google Scholar]
  29. Paquola, C., Bennett, M. R., Hatton, S. N., Hermens, D. F., & Lagopoulos, J. (2017). Utility of the cumulative stress and mismatch hypotheses in understanding the neurobiological impacts of childhood abuse and recent stress in youth with emerging mental disorder. Human Brain Mapping, 38(5), 2709–2721. 10.1002/hbm.23554 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Penza, K. M., Heim, C., & Nemeroff, C. B. (2003). Neurobiological effects of childhood abuse: Implications for the pathophysiology of depression and anxiety. Archives of Women's Mental Health, 6(1), 15–22. 10.1007/s00737-002-0159-x [DOI] [PubMed] [Google Scholar]
  31. Power, J. D., Cohen, A. L., Nelson, S. M., Wig, G. S., Barnes, K. A., Church, J. A., Vogel, A., Laumann, T., Miezin, F., Schlaggar, B., & Petersen, S. (2011). Functional network organization of the human brain. Neuron, 72(4), 665–678. 10.1016/j.neuron.2011.09.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Rakesh, D., Allen, N. B., & Whittle, S. (2023). Longitudinal changes in within-salience network functional connectivity mediate the relationship between childhood abuse and neglect, and mental health during adolescence. Psychological Medicine, 53(4), 1552–1564. 10.1017/S0033291721003135 [DOI] [PubMed] [Google Scholar]
  33. Schaefer, A., Kong, R., Gordon, E. M., Laumann, T. O., Zuo, X. N., Holmes, A. J., Eickhoff, S. B., & Yeo, B. T. T. (2018). Local-global parcellation of the human cerebral cortex from intrinsic functional connectivity MRI. Cerebral Cortex, 28(9), 3095–3114. 10.1093/cercor/bhx179 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Seghete KL, M., DePrince, A. P., & Banich, M. T. (2018). Association between initial age of exposure to childhood abuse and cognitive control: Preliminary evidence. Journal of Traumatic Stress, 31(3), 437–447. 10.1002/jts.22290 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Shao, X., Sun, D., Zhang, B., Cheng, L., Yan, C., & Zhu, G. (2020). Association between GPx-1 polymorphisms and personality traits in healthy Chinese-Han subjects. Brain and Behavior, 10(12), e01897. 10.1002/brb3.1897 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Sinha, R., Lacadie, C. M., Constable, R. T., & Seo, D. (2016). Dynamic neural activity during stress signals resilient coping. Proceedings of the National Academy of Sciences, 113(31), 8837–8842. 10.1073/pnas.1600965113 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Stocker, J. E., Nozari, E., van Vugt, M., Jansen, A., & Jamalabadi, H. (2023). Network controllability measures of subnetworks: Implications for neurosciences. Journal of Neural Engineering, 20(1), 016044. 10.1088/1741-2552/acb256 [DOI] [PubMed] [Google Scholar]
  38. Tymofiyeva, O., Hu, R., Kidambi, R., Nguyen, C., Max, J. E., & Yang, T. T. (2022). A meta-analysis of brain morphometric aberrations in adolescents who experienced childhood trauma. Frontiers in Human Neuroscience, 16, 1022791. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Wang, X., Ding, F., Cheng, C., He, J., Wang, X., & Yao, S. (2022). Psychometric properties and measurement invariance of the childhood trauma questionnaire (short form) across genders, time points and presence of major depressive disorder Among Chinese adolescents. Frontiers in Psychology, 13, 816051. 10.3389/fpsyg.2022.816051 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Wang, X., Wang, Y., & Xin, T. (2020). The psychometric properties of the Chinese version of the beck depression inventory-II With middle school teachers. Frontiers in Psychology, 11, 548965. 10.3389/fpsyg.2020.548965 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zhang, H., Wong, T. Y., Broekman, B. F. P., Chong, Y. S., Shek, L. P., Gluckman, P. D., Tan, K. H., Meaney, M. J., Fortier, M. V., & Qiu, A. (2021). Maternal adverse childhood experience and depression in relation with brain network development and behaviors in children: A longitudinal study. Cerebral Cortex, 31(9), 4233–4244. 10.1093/cercor/bhab081 [DOI] [PubMed] [Google Scholar]
  42. Zhang, Y., Du, X., Fu, Y., Zhao, Q., Wang, Z., Qin, W., & Zhang, Q. (2022). Effects of polygenic risk score of type 2 diabetes on the hippocampal topological property and episodic memory. Brain Imaging and Behavior, 16(6), 2506–2516. 10.1007/s11682-022-00706-w [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplementary materials.docx
ZEPT_A_2551953_SM1770.docx (43.9KB, docx)

Data Availability Statement

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Articles from European Journal of Psychotraumatology are provided here courtesy of Taylor & Francis

RESOURCES