Trauma Timing and Its Impact on Brain Activation During Flexible Emotion Regulation in PTSD: Insights From Functional MRI

ABSTRACT

Patients with Post‐Traumatic Stress Disorder (PTSD) exhibit deficits in flexible emotion regulation and display abnormal brain activation patterns. Previous research has not examined how the age at which trauma occurs influences associated behavioral and neural abnormalities. In this study, 76 adult participants (60.5% women) diagnosed with PTSD were categorized into three age‐matched groups based on the age at trauma onset: childhood, adolescence, and adulthood. Forty‐five healthy adults served as a control group. All participants engaged in the Shifted Attention Emotion Appraisal Task (SEAT) while undergoing functional magnetic resonance imaging (fMRI). Our findings reveal that both the childhood and adulthood trauma groups showed significantly greater activation in the left thalamus, left frontal gyrus, and Brodmann Area 48 compared to the adolescent trauma group. Additionally, the childhood trauma group exhibited higher activation in the left inferior frontal gyrus than the adolescent group and greater activation in the left pregenual anterior cingulate cortex compared to the adulthood trauma group. These results highlight the critical role of trauma timing in understanding the behavioral and neural dimensions of PTSD, offering new insights for clinical intervention and treatment strategies.

Both the childhood and adulthood trauma groups showed significantly greater activation in the left thalamus, left frontal gyrus, and Brodmann Area 48 compared to the adolescent group during emotion regulation by attention shifting. Childhood and adulthood trauma have a more profound impact on emotion regulation neural systems.

Summary.

• A retrospective case‐control design reveals novel insights into PTSD behavioral and neural abnormalities through the trauma age perspective.

• Childhood and adulthood trauma has a more profound impact on emotion regulation neural systems.

• Based on PTSD's complexity, fMRI objectively reveals implicit emotion regulation biomarkers, and the findings offer new insights for tailored PTSD clinical interventions and treatment strategies.

Abbreviations

ACC

anterior cingulate cortex

BA

Brodmann area

DLPFC

dorsolateral prefrontal cortex

FD

framewise Displacement

FDR

false discovery rate

fMRI

functional magnetic resonance imaging

FOV

field of view

GAD‐7

Generalized Anxiety Disorder 7 scale

GRF

Gaussian random field

KE

cluster size (voxels)

L

left

LFG

left fusiform gyrus

LHPC

left hippocampus

LIFG

left inferior frontal gyrus

LIFO

left inferior frontal operculum

L‐MCC

left mid‐cingulate cortex

LMFG

left middle frontal gyrus

LpACC

left pregenual anterior cingulate cortex

LPcu

left precuneus

LPreCG

left precentral gyrus

LROL

left rolandic operculum

LSFG

left superior frontal gyrus

LSMFG

left superior medial frontal gyrus

LSPL

left superior parietal lobule

LThal_MGN

left thalamus medial geniculate nucleus

LThal_VA

left thalamus ventral anterior nucleus

MNI

Montreal Neurological Institute

PCL‐5

Posttraumatic Stress Disorder Checklist for DSM‐5

PHQ‐9

Patient Health Questionnaire‐9 scale

PTSD

Post‐Traumatic Stress Disorder

R

right

RC

right caudate

RHPC

right hippocampus

RLG

right lingual gyrus

ROI

region of interest

RPcu

right precuneus

RPu

right putamen

RSFG

right superior frontal gyrus

RSOG

right superior occipital gyrus

SEAT

Shifted Attention Emotion Appraisal Task

SMA

supplementary motor area

SPM12

Statistical Parametric Mapping 12

TE

echo time

TR

repeat time

1. Introduction

Post‐Traumatic Stress Disorder (PTSD) is a chronic neuropsychiatric disorder triggered by exposure to traumatic events such as combat, sexual assault, natural disasters, or other severe accidents. The ramifications of PTSD extend beyond immediate psychological distress and include long‐term consequences. Individuals with PTSD frequently experience persistent symptoms, including intrusive memories, flashbacks, avoidance of trauma‐related stimuli, negative alterations in mood and cognition, and heightened arousal (American Psychiatric Association 2013). These symptoms can significantly impair daily functioning, disrupt relationships, and elevate the risk of comorbid conditions such as depression, substance use disorders, and other anxiety disorders (Sareen et al. 2007; Pietrzak et al. 2011). Survival analysis indicates that more than one‐third of those with an initial episode of PTSD do not recover even after many years (Kessler 1995). A meta‐analysis estimates that the lifetime prevalence of PTSD in the general population ranges from 5% to 10% (Ozer et al. 2003). The disorder is associated with substantial healthcare costs and productivity loss (Kessler 1995).

Emotional regulation is a key focus in PTSD research. Individuals with PTSD often struggle to manage their emotional responses, leading to increased emotional reactivity and difficulties in handling distress, and impairments in interpersonal relationships (Tull et al. 2007). Emotion regulation flexibility, the ability to adapt emotion regulation strategies to different situations, is crucial for mental health and well‐being. Studies have shown that greater emotion regulation flexibility is associated with lower levels of negative affect and higher levels of positive affect (Aldao et al. 2015). Attentional processes play a crucial role in emotion regulation by directing cognitive resources toward relevant stimuli and facilitating adaptive responses to environmental challenges. Individuals with PTSD often exhibit alterations in attentional processes, particularly in their ability to shift attention between different stimuli or tasks (Pineles et al. 2009). Previous studies have highlighted deficits in attentional control and attentional bias toward threat‐related stimuli in PTSD. Attentional bias refers to the tendency to selectively attend to threatening or trauma‐related cues while neglecting neutral or positive stimuli (Pineles et al. 2009; Bardeen et al. 2013). This bias can heighten emotional arousal and interfere with effective emotion regulation strategies, thereby contributing to the maintenance of PTSD symptoms. Therefore, emotion dysregulation is not only a consequence of PTSD but also a factor that perpetuates and exacerbates its symptoms over time (Eftekhari et al. 2009).

Traditionally, research on PTSD has primarily focused on explicit emotions, which are consciously experienced and self‐reported by individuals. This approach, however, may not fully capture the complex emotional disturbances inherent to PTSD. Limitations of using explicit emotional measures include the reliance on self‐report measures, which are susceptible to subjective biases such as social desirability and cognitive distortions, potentially compromising the accuracy of emotional evaluations (Foa and Kozak 1986). Individuals with PTSD may struggle to articulate their emotional experiences due to avoidance behaviors or emotional numbing, thereby compromising the reliability of their reports (Etkin and Wager 2007). Furthermore, explicit measures often fail to detect automatic, non‐conscious emotional responses, which are crucial for a comprehensive understanding of emotional dysregulation in PTSD (Pineles et al. 2009).

In contrast, implicit emotions, which arise automatically and unconsciously, provide valuable insights for understanding PTSD. First, implicit emotions are assessed through objective measures such as physiological indicators and behavioral responses, providing more reliable insights into emotional processing without the biases inherent in subjective self‐report methods (Wessa and Flor 2007). Second, these measures unveil emotional reactions that individuals may not consciously recognize or acknowledge, thereby elucidating the full spectrum of emotional dysregulation in PTSD that may be obscured by explicit measures (Jovanovic and Norrholm 2011). The Shifted Attention Emotion Appraisal Task (SEAT; Anderson et al. 2013; Klumpp et al. 2011; Duval et al. 2020) used in this study has emerged as a valuable tool for studying implicit emotion in PTSD, offering insights into emotional processing that may not be fully captured by traditional explicit measures (Liberzon et al. 2015; Duval et al. 2018). By studying physiological and behavioral markers, researchers can identify biomarkers of PTSD severity, predict treatment outcomes, and develop interventions targeting the automatic emotional processes that underpin PTSD symptoms.

This study investigates the onset age of PTSD and its neural consequences. According to widely accepted international age classifications, childhood ranges from birth to 12 years, adolescence from approximately 13–17 years, and adulthood from 18 years onward (Dick and Ferguson 2015; DunnGalvin et al. 2008; Zhu et al. 2020). These classifications are adopted for this analysis. The severity and manifestation of PTSD are influenced by many factors, including current age and age at trauma onset. Numerous studies have highlighted that individuals across different age groups exhibit distinct presentations of PTSD. Children with PTSD typically manifest symptoms such as nightmares, hypervigilance, and separation anxiety (Connor et al. 2015). Neurobiologically, PTSD in children is linked to alterations in brain structures important for memory and emotion regulation, such as the hippocampus and anterior cingulate cortex (Rinne‐Albers et al. 2017). Adolescents experiencing PTSD frequently exhibit emotional dysregulation, irritability, and social withdrawal (Paulus et al. 2021). Neuroimaging studies suggest that adolescent PTSD is associated with disrupted connectivity in brain regions implicated in emotional control and stress response, reflecting ongoing neurodevelopmental changes (Thomason et al. 2015; Xu et al. 2018). Adults with PTSD often present with symptoms such as hyperarousal, avoidance behaviors, and intrusive memories (Briere et al. 2015). Neurobiologically, adults show alterations in brain regions involved in fear processing and emotional regulation, including the amygdala and prefrontal cortex, as evidenced by functional magnetic resonance imaging (fMRI) studies (Alexandra Kredlow et al. 2022). While considerable research has explored the effects of present age on PTSD outcomes, less attention has been given to the effect of the age at trauma exposure. Previous studies of brain mechanisms in PTSD patients have predominately been cross‐sectional, with a lack of retrospective case–control studies that explore unique biomarkers across age groups and trace differences in the causes of symptoms.

The differences in PTSD among different age groups confound whether this is due to the participants' present age or the age at which the trauma was experienced. Adults with PTSD may have experienced trauma in childhood or adolescence, adolescents with PTSD often experience trauma in childhood, and children with PTSD usually experience trauma in early childhood. To address this, we categorized participants according to the age of trauma exposure, analyzed the effects on brain mechanisms and behavior, and explored the impact of different trauma exposure ages on these variables. This approach enables the identification of biomarkers unique to each age group, enhances the prediction of PTSD susceptibility based on trauma duration, refines diagnostic criteria to better capture symptom diversity, and facilitates the development of personalized treatment strategies addressing individual differences in emotional regulation and attention processing.

2. Methods

The study was approved by the Ethics Institutional Review Board and was registered with the Chinese Clinical Trial Registry (ChiCTR2200058408). The research process is shown in Figure 1.

FIGURE 1.

Flowchart of the empirical study procedure.

2.1. Participants

One hundred and thirty‐one participants aged 18–45 years were recruited for the present study via the Internet and hospital psychiatric clinics, and were financially compensated for participation. Exclusion criteria included a history of psychosis, major depressive disorder, and other psychiatric conditions, significant medical conditions, neurological illnesses, or a history of head injury. Written informed consent was obtained from all participants.

All patients with PTSD had experienced traumatic events and the score of the PTSD Symptom Screening Scale was higher than 31 points. Clinical symptoms such as anxiety and depression scores were assessed using the Generalized Anxiety Disorder 7 scale (GAD‐7; Ahmad et al. 2017) and the Patient Health Questionnaire‐9 (PHQ‐9; Kroenke and Spitzer 2002). Diagnoses and interview assessments were performed by trained psychiatric clinicians. After excluding three participants in the healthy control group and seven in the PTSD group due to poor technical performance, the remaining 45 healthy controls (M _age = 21.38, SD_age = 3.40, 28 female) and 76 PTSD patients (M _age = 21.79, SD_age = 3.85, 46 female) completed fMRI scans. The whole PTSD sample was assigned to three subgroups based on the age of trauma occurrence. Seventeen patients (M _age = 22.53, SD_age = 4.14, 10 female) experienced trauma in childhood (below 12 years old), 29 patients (M _age = 20.48, SD_age = 2.61, 18 female) experienced trauma in adolescence (between 13 and 17 years old), and 30 patients (M _age = 22.63, SD_age = 4.42, 18 female) experienced trauma in adulthood (above 18 years old).

Statistical analyses were conducted using SPSS 27.0. The chi‐square test and Monte Carlo simulation (When the expected frequency was < 5) exact test were utilized to analyze categorical variables such as sex, region, and education. For continuous variables PCL‐5 (including the four symptoms of PTSD: intrusion, avoidance, negative alterations in cognitions and mood, and alterations in arousal and reactivity), GAD‐7, and PHQ‐9, normality of distribution was assessed using the Shapiro–Wilk test, and homogeneity of variances was evaluated using Levene's test. Comparisons between the All HC group and the All PTSD group utilized independent samples t‐tests for variables satisfying the assumptions of normality and homogeneity of variance; otherwise, Mann–Whitney U tests were employed. For comparisons among the three PTSD subgroups (childhood/adolescence/adulthood trauma), one‐way ANOVA was applied when normality and homogeneity assumptions were met, with Bonferroni‐adjusted pairwise comparisons conducted for significant effects. When these assumptions were violated, the Kruskal‐Wallis H test was used; upon significant results (p < 0.05), post hoc pairwise comparisons were performed using Mann–Whitney U tests with Bonferroni correction.

2.2. Procedure

Data were collected from April 2023 to April 2024. All participants completed the SEAT task during fMRI scanning. See Figure 2a. The SEAT task presents compound stimuli that include both emotional faces and neutral scenes (Ekman and Friesen 1976). The stimuli include 60 composite pictures of superimposed faces (foreground) and buildings (background), as well as 20 pictures of faces or buildings only. The face pictures depict neutral, angry, or fearful expressions, and the building pictures depict indoor or outdoor scenes. Under three different conditions, participants were asked to respond to the following questions: (1) ‘Gender’: whether the face in the foreground is male or female; (2) ‘Inside/Outside’: whether the scene in the background is indoors or outdoors; (3) ‘Like/Dislike’: whether the face in the foreground is liked or disliked; (4) ‘Face/Place’: whether the image is a face or a place.

FIGURE 2.

(a) Examples of stimuli from the Shifted Attention Emotion Appraisal Task (SEAT). (b) Brain clusters showing significant positive or negative activation in patients with PTSD compared to healthy controls (HC) during emotion regulation by attention shifting (Indoor/Outdoor–Male/Female) in the SEAT paradigm. (c) Differences in task‐related activations between the PTSD group and HC participants, highlighting the distinct neural responses during emotional regulation.

The ‘Gender’ condition probes implicit emotional processing, as attention is focused on an emotional face while one is identifying its gender, and it has been well established that negative emotional faces induce corresponding negative emotions (Dyck et al. 2011; Schneider et al. 1997; Schneider et al. 1994). The ‘Inside/Outside’ condition engages emotion regulation by attention shifting, as it requires focusing attention on the building components superimposed on the emotional face. Attention modulation is recognized as a core component of emotion regulation (Gross 2013), and attentional modulation away from threatening faces lowers frustration for subsequent stressful tasks (Johnson 2009). The ‘Like/Dislike’ condition engages cognitive appraisal of one's emotional/evaluative state while processing the emotional stimulus. The ‘Face/Place’ controls for brain activation associated with simply viewing faces and scenes.

Sixty composite pictures were each presented three times (E‐Prime Inc., Sharpsburg, PA), once for each condition, with condition types presented in random order. The non‐composite pictures (showing only a face or scene) were presented in 40 trials total. There were four runs, with 55 trials per run. Trials comprised a centered fixation crosshair for 3–8 s, a judgment cue for 750 + 250 ms blank screen, and a composite image for 1500 ms. Prior to the experimental trials, participants completed a practice session with images not used in the experiment. The entire task took approximately 32 min.

2.3. Self‐Report Measures

2.3.1. Post‐Traumatic Stress Disorder

The Posttraumatic Stress Disorder Checklist‐5 (PCL‐5; Blevins et al. 2015) was utilized to assess the severity of PTSD symptoms. The PCL‐5 demonstrates high internal consistency reliability (Cronbach's α = 0.94), excellent discriminant validity, and high sensitivity to clinical change (Weathers 2017). In this study, participants with a PCL‐5 score greater than 31 were included. The PCL‐5 total score evaluated the four core symptoms of PTSD: intrusion, avoidance, negative alterations in cognitions and mood, and hyperarousal. A higher score on the PCL‐5 indicates a greater severity of trauma‐related symptoms.

2.3.2. Anxiety

Anxiety levels were measured using the Generalized Anxiety Disorder 7 (GAD‐7) scale, which exhibits an internal consistency reliability of Cronbach's α = 0.92 (Ahmad et al. 2017). A higher score is indicative of increased anxiety levels.

2.3.3. Depression

Depression symptoms were assessed with the Patient Health Questionnaire‐9 (PHQ‐9; Kroenke and Spitzer 2002). The PHQ‐9 has an internal consistency reliability of Cronbach's α = 0.89 (Kroenke et al. 2001). A higher score means more severe symptoms. All participants completed these scales.

2.4. MRI Data Acquisition

Scanning was performed on a 3T Philips scanner at Wuhan Psychological Hospital of China. Structural images were obtained using a magnetization prepared rapid gradient echo T1‐weighted sequence with the following scanning parameters: Repeat time (TR)/echo time (TE) = 1900/2.48 ms, matrix = 256 × 256, field of view (FOV) = 256 mm × 256 mm, flip angle = 9°, thickness/gap = 1/0.5 mm. Functional images were collected by gradient‐echo planar imaging sequence. The scanning parameters were: TR/TE = 2000/30 ms, matrix = 64 × 64, FOV = 240 mm × 240 mm, turning angle = 90°, layer thickness/interval = 4/1 mm. A total of 100 time points were acquired. E‐Prime software (Psychology Software Tools, Pittsburgh, PA) was used to present stimuli and record responses. During the task‐state scan, participants watched on‐screen stimuli and responded using an MRI‐compatible button box.

2.5. Whole Brain Voxel‐Wise fMRI Analyses

For the fMRI analyses, three participants were excluded due to movements exceeding 3 mm. Additionally, seven participants were excluded due to not finishing the task, technical problems, or artifacts during data collection. Therefore, the final sample for the fMRI analyses included 121 participants.

All MRI data analyses were performed using Statistical Parametric Mapping (SPM12; Welcome Centre for Human Neuroimaging, London, UK) for MATLAB, with smoothing applied using an 8 mm kernel. All motion parameters and their derivatives, along with regressors for motion outlier time points (defined as framewise displacement > 0.5 mm or standardized DVARS > 1.5), were included as nuisance regressors in the subject‐level analysis. The results presented in this manuscript are derived from preprocessing performed using fMRIPrep 23.1.4 (Esteban et al. 2018, 2019; RRID: SCR_016216), based on Nipype 1.8.6 (Gorgolewski et al. 2011, 2018; RRID: SCR_002502). See Appendix S1 for details.

Implicit emotion regulation involves both attention shifting and appraisal components. To isolate brain circuits related to the different emotion regulation processes discussed, we created the following three specific contrasts in the first‐level analysis for each participant: (1) Male/Female—Face/Place Only, (2) Indoor/Outdoor—Male/Female, and (3) Like/Dislike—Male/Female. Contrast images from the first‐level comparisons were carried forward to the second‐level analyses. (a) Firstly, a one‐sample t test was conducted on the whole‐brain data of all participants (N = 121) to identify which brain regions are activated by the SEAT paradigm. (b) Next, a two‐sample t test was performed on the whole‐brain data comparing PTSD group and HC group to investigate differences in implicit emotion regulation between PTSD and healthy individuals. (c) Finally, analyses were conducted on the whole‐brain data comparing PTSD subgroups stratified by trauma exposure timing (childhood/adolescent/adulthood trauma exposure). To identify regions showing significant main effects of trauma exposure subgroup, a voxel‐wise one‐way ANOVA was first conducted among the three PTSD subgroups. Specific pairwise differences between subgroups (childhood vs. adolescence; childhood vs. adulthood; adolescence vs. adulthood) were subsequently assessed using direct two‐sample t tests.

For all analyses (a–c), age, sex, and mean framewise displacement (FD) were included as covariates of no interest. A stricter false discovery rate (FDR) correction was applied for the one‐sample t test (voxel‐wise p < 0.001, cluster‐wise p < 0.05). For the two‐sample t tests (b: PTSD vs. HC; c: pairwise subgroup comparisons), Gaussian random field (GRF) theory correction was applied to control family‐wise error, with a threshold of voxel‐wise p < 0.001 or 0.005 and cluster‐wise p < 0.05, following the recommendations of Eklund and colleagues (Eklund et al. 2016; Xie et al. 2022). All statistical analyses were conducted using the DPABI toolbox (Yan et al. 2016) in MATLAB. Significant activation clusters were localized and reported using the Brodmann atlas and the Automated Anatomical Labeling (AAL3) template.

2.6. Region‐of‐Interest Analyses

Seven ROIs were selected based on attention shifting (Indoor/Outdoor—Male/Female) activation from the one‐sample t‐test results of the PTSD subjects (N = 76, see Table S1 and Figure 3b). All regions were defined as 5 mm radius spheres around the peak coordinates, as recommended by Masharipov et al. (2024) and Harvey et al. (2018). Beta weights were extracted using the DPABI in MATLAB. Data were analyzed using SPSS 27.0.

FIGURE 3.

(a) Brain activation results of two‐sample t‐tests for pairwise comparison among three subgroups during emotional regulation by attention shifting (Indoor/Outdoor–Male/Female) in the SEAT paradigm. From top to bottom are: differences between the childhood trauma group and the adolescent trauma group, differences between the adulthood trauma group and the adolescent trauma group, differences between the childhood trauma group and the adulthood trauma group. (b) Brain clusters exhibiting significant activation changes in all PTSD patients during emotion regulation by attention shifting in the SEAT task.

Shapiro–Wilk tests were conducted to assess the normality of ROI beta values and clinical measures (PCL‐5, GAD‐7, PHQ‐9). Given the presence of covariates (age, sex, trauma age and mean FD) in our analyses, exploratory partial correlation analyses were employed to control for these variables. Specifically, Pearson partial correlation was used when both variables met the normality assumptions; otherwise, Spearman's rho partial correlation was applied. This approach aimed to identify ROIs significantly associated with clinical symptoms across PTSD.

3. Results

3.1. Demographics and Symptoms

Chi‐square tests and Monte Carlo simulation exact tests revealed no significant differences between the All PTSD and All HC groups in age, gender, residence, education level, marital status, working status, or income satisfaction (all p > 0.05). Similarly, comparisons among the three trauma‐onset groups (childhood, adolescence, adulthood) showed no significant differences in these demographic variables (all p > 0.05; see Table 1).

TABLE 1.

Demographic and clinical data of participants.

Variables	ALL HC (n = 45)		ALL PTSD (n = 76)		p	Trauma experienced when childhood (n = 17)		Trauma experienced when adolescence (n = 29)		Trauma experienced when adulthood (n = 30)		Analysis of variance
	Mean/N	SD (%)	Mean (N)	SD (%)		Mean (N)	SD (%)	Mean (N)	SD (%)	Mean (N)	SD (%)	χ 2 (H/F)	df	p
Demographic characteristics
Age (years)	21.38	3.40	21.79	3.85	0.49	22.53	4.14	20.48	2.61	22.63	4.42	2.84	2	0.07
Sex	—	—	—	—	1.00	—	—	—	—	—	—	0.12	2	1.00
Male	17	37.78%	30	39.47%	—	7	41.18%	11	37.93%	12	40.00%	—	—	—
Female	28	62.22%	46	60.53%	—	10	58.82%	18	62.07%	18	60.00%	—	—	—
Residence	—	—	—	—	1.00	—	—	—	—	—	—	4.65	4	0.28
Urban	37	82.22%	60	78.95%	—	11	64.71%	23	79.31%	26	86.67%	—	—	—
Rural	7	15.56%	13	17.11%	—	4	23.53%	5	17.24%	4	13.33%	—	—	—
Suburb	1	2.22%	3	3.95%	—	2	11.76%	1	3.45%	—	—	—	—	—
Education level	—	—	—	—	0.88	—	—	—	—	—	—	7.69	6	0.18
High or vocational school and below	—	—	2	2.63%	—	1	5.88%	1	3.45%	—	—	—	—	—
Junior college	3	6.67%	5	6.58%	—	1	5.88%	2	6.90%	2	6.67%	—	—	—
Undergraduate college	37	82.22%	62	81.58%	—	12	70.59%	26	89.66%	24	80.00%	—	—	—
Master and above	5	11.11%	7	9.21%	—	3	17.65%	—	—	4	13.33%	—	—	—
Marital status	—	—	—	—	0.96	—	—	—	—	—	—	5.69	6	0.43
Married and live with spouse	2	4.44%	3	3.95%	—	1	5.88%	—	—	2	6.67%	—	—	—
Married but not live with spouse	12	26.67%	18	23.68%	—	4	23.53%	5	17.24%	9	30.00%	—	—	—
Single	31	68.89%	54	71.05%	—	12	70.59%	24	82.76%	18	60.00%	—	—	—
Unmarried and in a stable relationship	—	—	1	1.32%	—	—	—	—	—	1	3.33%	—	—	—
Working status	—	—	—	—	—	—	—	—	—	—	—	4.45	4	0.30
Full‐time	4	8.89%	5	6.58%	0.23	2	11.76%	—	—	3	10.00%	—	—	—
Part‐time	1	2.22%	—	—	—	—	—	—	—	—	—	—	—	—
Student	40	88.89%	67	88.16%	—	15	88.24%	27	93.10%	25	83.33%	—	—	—
Unemployed	—	—	4	5.26%	—	—	—	2	6.90%	2	6.67%	—	—	—
Income satisfaction	—	—	—	—	—	—	—	—	—	—	—	5.67	6	0.46
Completely satisfied	10	22.22%	10	13.16%	0.54	3	17.65%	5	17.24%	2	6.67%	—	—	—
Somewhat satisfied	25	55.56%	51	67.11%	—	11	64.71%	19	65.52%	21	70.00%	—	—	—
Not very satisfied	6	13.33%	9	11.84%	—	1	5.88%	2	6.90%	6	20.00%	—	—	—
Not at all satisfied	4	8.89%	6	7.89%	—	2	11.76%	3	10.34%	1	3.33%	—	—	—
Clinical characteristics
PCL‐5	—	—	43.11	10.07	—	44.65	9.91	39.48	9.50	45.73	9.95	6.31	2	0.043*
Cluster B	—	—	11.83	3.69	—	11.53	3.26	10.55	3.98	13.23	3.20	4.33	2	0.017*
Cluster C	—	—	4.29	2.20	—	3.82	2.38	3.90	1.95	4.93	2.24	4.56	2	0.10
Cluster D	—	—	15.36	4.78	—	16.88	3.90	14.48	5.22	15.33	4.72	3.61	2	0.16
Cluster E	—	—	11.63	3.62	—	12.41	3.71	10.55	3.10	12.23	3.89	2.16	2	0.12
GAD‐7	—	—	11.07	4.63	—	11.76	4.09	9.97	4.49	11.73	4.99	2.62	2	0.27
PHQ‐9	—	—	13.70	5.02	—	13.88	5.22	13.00	4.36	14.27	5.57	0.48	2	0.62

Note: The data are presented as the mean ± standard deviation. Significance values are based on two‐tailed tests, “*”, p < 0.05. Gender: 1 = male, 2 = female; Residence: 1 = urban, 2 = rural, 3 = suburb; Education level: 1 = high or vocational school and below, 2 = junior college, 3 = undergraduate, 4 = college, 5 = master and above; Marital status: 1 = married and live with spouse, 2 = married but not live with spouse, 3 = single, 4 = unmarried and in a stable relationship; working status: 1 = full‐time, 2 = part‐time, 3 = student, 4 = unemployed; Income satisfaction: 1 = completely satisfied, 2 = somewhat satisfied, 3 = not very satisfied, 4 = not at all satisfied.

Abbreviations: ALL HC, all healthy control; ALL PTSD, all PTSD participants; Cluster B, intrusion; Cluster C, avoidance; Cluster D, negative alterations in cognitions and mood; Cluster E, hyperarousal; F, results of one‐way ANOVA; GAD‐7, Generalized Anxiety Disorder 7 scale; H, the result of Kruskal‐Wallis ANOVA; PCL‐5, Posttraumatic Stress Disorder Checklist for DSM‐5; PHQ‐9, Patient Health Questionnaire‐9 scale; SD standard deviation; χ ², the result of chi‐square test or Monte Carlo simulation exact tests.

The Shapiro–Wilk test revealed non‐normal distributions for PTSD Checklist‐5 (PCL‐5) total score (Group 2: p = 0.02; Group 3: p = 0.02), negative alterations in cognition and mood (Cluster C; Group 3: p = 0.03), alterations in arousal and reactivity (Cluster D; Group 1: p = 0.03), and Generalized Anxiety Disorder‐7 (GAD‐7; Group 1: p = 0.03). In contrast, intrusion symptoms (Cluster B; all p > 0.21), avoidance symptoms (Cluster E; all p > 0.25), and Patient Health Questionnaire‐9 (PHQ‐9; all p > 0.05) satisfied normality assumptions across all groups (see Table S2).

Kruskal‐Wallis analysis demonstrated statistically significant differences among trauma‐onset groups in total PTSD symptom severity (PCL‐5; H = 6.31, p = 0.043). As detailed in Table S3, post hoc analysis revealed a trend toward higher scores in the adult‐onset group compared to the adolescent‐onset group (z = −2.37, p = 0.054). Childhood‐onset PTSD severity did not significantly differ from other groups (both p > 0.22).

Homogeneity of variances was confirmed for intrusion symptoms (Cluster B) using Levene's test (F (2, 73) = 0.90, p = 0.41). One‐way ANOVA revealed significant differences among trauma‐onset groups (F (2, 73) = 4.33, p = 0.017, η ²  = 0.11). Post hoc Bonferroni tests indicated significantly lower intrusion symptoms in the adolescent‐onset group compared to the adult‐onset group (mean difference = −2.68, p = 0.014). Complete pairwise comparison results are presented in Table S4.

For avoidance symptoms (Cluster E), Levene's test confirmed homogeneity of variances (F (2, 73) = 1.18, p = 0.31). Similarly, homogeneity of variances was established for depression (PHQ‐9; F (2, 73) = 1.79, p = 0.17). One‐way ANOVA indicated no significant differences for avoidance symptoms (Cluster E; F (2, 73) = 2.16, p = 0.12) or depression (PHQ‐9; F (2, 73) = 0.48, p = 0.62). Similarly, Kruskal‐Wallis tests showed no significant differences for negative alterations in cognition and mood (Cluster C; H = 4.56, p = 0.10), alterations in arousal and reactivity (Cluster D; H = 3.61, p = 0.16), or generalized anxiety (GAD‐7; H = 2.62, p = 0.27).

3.2. fMRI Task‐Related Activations

We used the Male/Female condition (implicit emotion processing) as a baseline to examine the effects of emotion regulation by attention shifting (Indoor/Outdoor). The one‐sample t‐test on fMRI data from 121 participants revealed significant positive activation in the LFG, LMFG, and RSFG. Negative activation was observed in the RHPC, LHPC, LSMFG, and L‐MCC. See Figure 2b and Table 2. A two‐sample t‐test comparing fMRI data during emotion regulation by attention shifting between the PTSD and HC groups revealed lower activation in the PTSD group in the LFG, RLG, RPcu, LSPL, and RSOG. Conversely, the PTSD group exhibited significantly higher activation in the LPcu (see Figure 2c and Table 2).

TABLE 2.

Neural activation patterns during emotion regulation by attention shifting: within‐group task activation and between‐group differences (PTSD vs. HC).

Region		MNI coordinates			KE	t
		x	y	z
One‐sample t test in all participants during emotion regulation by attention shifting a
Positive activation	L fusiform gyrus (LFG, BA37)	−30	−38	−15.5	3873	12.8
	L middle frontal gyrus (LMFG)	−26.5	11	51	423	7.07
	R superior frontal gyrus (RSFG)	26	11	54.5	145	6.34
Negative activation	R hippocampus (RHPC)	19	−6.5	−12	42	−5.6
	L hippocampus (LHPC)	−19.5	−10	−12	30	−4.88
	L superior medial frontal gyrus (LSMFG)	−5.5	53	23	339	−7.77
	L mid‐cingulate cortex (L‐MCC)	−2	−17	33.5	47	−5.06
Whole PTSD group > HC group (two‐sample t test) during emotion regulation by attention shifting b
Positive activation	L Precuneus (LPcu)	−5.5	−59	30	92	3.74
Negative activation	L fusiform gyrus (LFG, BA37)	−23	−48.5	−12	125	−3.88
	R lingual gyrus (RLG)	29.5	−45	−8.5	67	−3.72
	R Precuneus (RPcu, BA29)	8.5	−48.5	5.5	37	−3.15
	L superior parietal lobule (LSPL, BA7)	−19.5	−73	51	160	−3.94
	R superior occipital gyrus (RSOG, BA19)	26	−80	44	90	−3.41

^a For one‐sample t tests on all participants under the three SEAT task conditions, a stricter false discovery rate (FDR) correction was applied (voxel‐wise p < 0.001, cluster‐wise p < 0.05).

^b For two‐sample t tests comparing groups, Gaussian random field (GRF) correction was used, with a voxel threshold of p < 0.005 and a cluster threshold of p < 0.05.

ANOVA analysis showed that emotion regulation by attention shifting (Indoor/Outdoor—Male/Female) activated peaks (−23, 25, and 5.5) in the left putamen (BA48) across three subgroups.

A between‐group comparison showed that, compared to adolescent trauma patients, childhood and adulthood trauma groups exhibited greater activations in the left thalamus regions (LThal_VA, LThal_MGN), left frontal gyrus regions (LSFG, LPreCG) and Brodmann Area 48 (BA48; involved in LIFO, LROL, RC, and RPu respectively) during attention modulation. Childhood trauma patients exhibited heightened activation in the left inferior frontal gyrus (LIFG, LIFO) compared to adolescents. In contrast, adulthood trauma patients showed decreased activation in the left pregenual anterior cingulate cortex (LpACC) compared to the childhood group (Figure 3a, Table 3).

TABLE 3.

Brain activation results of differences between PTSD patients with trauma exposure at different ages (two‐sample t test) during emotion regulation by attention shifting.

Region	MNI coordinates			K E	t	P GRF
	x	y	z
Childhood experienced>adolescence experienced
L inferior frontal gyrus (LIFG, BA45)	40	39	13	146	4.16	0.001
R caudate (RC, BA48)	−23	25	5.5	71	4.11	0.001
L thalamus ventral anterior nucleus (LThal_VA)	12	−3	9	64	4.03	0.001
L superior frontal gyrus (LSFG, BA46)	22.5	39	26.5	63	4.21	0.001
L inferior frontal operculum (LIFO, BA48)	50.5	11	9	35	4.38	0.001
Adulthood experienced>adolescence experienced
L rolandic operculum (LROL, BA48)	33	4	16	65	3.83	0.001
R putamen (RPu, BA48)	−26.5	18	5.5	34	3.77	0.001
L thalamus medial geniculate nucleus (LThal_MGN)	15.5	−24	−8.5	33	3.95	0.001
L precentral gyrus (LPreCG, BA6)	40	4	37	27	3.80	0.001
Childhood experienced>adulthood experienced
L pregenual anterior cingulate cortex (LpACC, BA32)	1.5	42.5	12.5	30	3.43	0.005

Note: Gaussian random field (GRF) correction was used, with a voxel threshold of p < 0.001 or 0.005 and a cluster threshold of p < 0.05.

Abbreviations: BA, Brodmann area; KE, cluster size (voxels); L, left; MNI, Montreal Neurological Institute; R, right; T, peak activation T value.

3.3. Regions of Interest Analysis

The Shapiro–Wilk test results for the ROIs in the whole PTSD samples can be found in Table S1. After FDR correction for multiple comparisons (21 tests across 7 ROIs and 3 clinical measures), no significant associations were observed between neural activity and clinical symptoms (all q > 0.05). Uncorrected results are provided in Table S5 for transparency.

4. Discussion

Our study systematically investigates how trauma timing affects emotion regulation deficits and neural activation in PTSD patients. Results revealed that both childhood and adulthood trauma groups exhibited significantly greater activation in the left thalamus, left frontal gyrus, and Brodmann Area 48 compared to the adolescent group. The childhood trauma group showed higher activation in the left inferior frontal gyrus and left pregenual anterior cingulate. These results suggest that the timing of trauma exposure in PTSD patients not only influences the degree of emotion regulation deficits but also potentially modulates neural activation across key brain regions involved in emotional and cognitive processing. These neural differences highlight the importance of considering trauma timing when examining emotion regulation and its underlying neurobiological mechanisms in PTSD.

4.1. Key Symptomatology Findings by Trauma Onset Timing

The absence of significant demographic differences (age, gender, socioeconomic factors) between the PTSD onset groups and HCs strengthens the validity of our subsequent symptom profile interpretations by eliminating these key potential confounds. Notably, our findings reveal a significant developmental sensitivity specifically for intrusion symptoms (Cluster B). Intrusion severity was significantly higher following trauma exposure in adulthood compared to adolescence (p = 0.014). This suggests that once neurocognitive maturation is largely complete, the brain may exhibit reduced flexibility in adaptively integrating traumatic experiences into existing cognitive schemas, potentially leading to a greater propensity for persistent, involuntary re‐experiencing of the trauma (Brewin 2014; Ehlers and Clark 2000). This interpretation is further supported by a trend (p = 0.054) towards higher overall PTSD symptom severity in the adult‐onset group, indicating a potentially greater overall symptom burden associated with trauma occurring after this developmental window, although replication of this specific finding is warranted.

Conversely, the childhood‐onset PTSD group did not exhibit distinguishing patterns in self‐reported symptom severity compared to adolescent or adult‐onset groups within this cohort. However, as detailed in subsequent neurobiological assessments, distinct developmental pathways associated with this very early trauma exposure were evident, highlighting a divergence between subjective symptom report and underlying neurobiological alterations that merits further exploration.

Additionally, the core symptom dimensions of avoidance, negative alterations in cognition and mood (dysphoria), and alterations in arousal and reactivity (hyperarousal), alongside comorbid depression and anxiety severity, showed no significant variation based on the developmental timing of the index trauma. This key finding implies that these pervasive aspects of post‐traumatic psychopathology are less sensitive to the specific neurodevelopmental stage at which the trauma occurs.

4.2. The Activation of SEAT

The SEAT paradigm is designed to investigate implicit emotion regulation through attention shifting, engaging a broad neural network that includes regions critical for both emotional processing and cognitive control (Ma et al. 2017). Our findings, which revealed significant activation in the LFG, LMFG, and RSFG, as well as deactivation in the RHPC, LHPC, LSMFG, and L‐MCC during attention shifting, align with previous studies utilizing the SEAT paradigm (Liberzon et al. 2015; Duval et al. 2018). Our results further underscore the role of these brain circuits in modulating automatic emotional responses, especially in PTSD, where abnormal functional connectivity in these regions has been well reported (Thome et al. 2020).

The negative activation in the LSMFG, associated with cognitive control, while positive activation in regions involved in facial and emotional processing, including the left and right fusiform gyrus (LFG, RFG). Increased fusiform gyrus activation reflects PTSD's impact on emotional processing (Liberzon and Martis 2006). These neural patterns tentatively align with hypotheses proposing that cognitive control mechanisms may be recruited during attention‐shifting tasks as a potential compensatory strategy for emotion regulation difficulties (Etkin and Wager 2007; Hayes et al. 2012; Zilverstand et al. 2017). Future studies with larger samples are needed to further investigate these mechanisms.

In PTSD patients, our results showed altered activation patterns compared to healthy controls, with lower activation in regions such as the LFG, RLG, RPcu, LSPL, and RSOG, and higher activation in the LPcu. These findings are in line with previous research indicating that PTSD is associated with abnormal functional connectivity in brain regions involved in emotional processing and regulation (Thome et al. 2020). The reduced activation in the left and right fusiform gyri (LFG and RLG) may reflect impairments in the processing of emotional stimuli, particularly in the recognition and evaluation of emotional information (Wang et al. 2008). The fusiform gyri are known to be involved in facial emotion recognition tasks, and reduced activation in these regions has been associated with deficits in emotional processing (Tang et al. 2017). In contrast, the increased activation in the left precuneus (LPcu) could suggest heightened self‐monitoring and attention allocation, possibly as a compensatory mechanism to maintain cognitive and emotional functioning despite the impairments in emotional processing (Wang et al. 2008). The precuneus is a key region in the default mode network (DMN) and is involved in self‐referential thought, introspection, and social cognition (Li et al. 2025). Increased activation in the LPcu has been observed in conditions where there is a need for heightened self‐monitoring and attention allocation (Wang et al. 2008). This pattern of activation highlights the dynamic interplay between regions involved in emotional processing and those engaged in cognitive control and self‐monitoring (Ferri et al. 2016).

The SEAT paradigm effectively engages brain regions associated with implicit emotion regulation by eliciting automatic emotional responses and modulating them through attention shifting. This is supported by the overlap between our findings and those from other SEAT studies, which link these regions to altered emotional processing and regulation (Barkay et al. 2012; Kunimatsu et al. 2020).

4.3. Trauma Timing and Neural Activation

The left thalamus, essential for sensory processing, motor control, and emotional regulation, plays a critical role in PTSD (Li et al. 2021; Wang et al. 2012). The LThal_MGN, involved in emotional processing and linked to the amygdala, is particularly relevant in fear and anxiety (Chen 2019; Leppla 2019). In PTSD patients, increased activation in the left thalamus during emotional regulation tasks reflects heightened sensitivity to sensory stimuli, indicating difficulties in disengaging from negative stimuli. Childhood and adulthood trauma groups exhibited greater left thalamus (LThal_VA/LThal_MGN) activation during emotional regulation tasks than the adolescent group, indicating more severe symptoms and emotion regulation difficulties. The differential activation in LThal_VA and LThal_MGN can be attributed to their varying roles at different developmental stages. The Thal_VA is crucial for motor function development (Metzger et al. 2013), particularly in early childhood. Trauma at this stage could increase Thal_VA activation, potentially altering long‐term brain function (Blennow et al. 1996), particularly in regions responsible for motor control and coordination (Piaget and Cook 1952). In adulthood, heightened Thal_MGN activation may signify constant alertness and difficulty distinguishing safe from dangerous sounds, contributing to trauma‐related symptoms (Williams 2021).

The frontal gyrus, particularly the dorsolateral prefrontal cortex (DLPFC, including LSFG) and supplementary motor area (SMA, included in LPreCG), is vital for cognitive and motor functions (Ma et al. 2017; Song et al. 2017). LSFG (BA46) is involved in top‐down modulation of emotional responses, while LPreCG (BA6) is responsible for planning and executing attention shifts (Phillips et al. 2003; Ochsner et al. 2002; Banks et al. 2007). Increased left frontal gyrus activation during attention‐shifting tasks indicates active cognitive engagement. However, it may also reflect emotion regulation difficulties, especially in trauma groups. Dysregulation of LSFG, a critical part of the DLPFC, may impair emotion regulation, contributing to anxiety and depressive rumination (Shapero et al. 2019; Wang et al. 2018).

Increased LSFG activation might reflect compensatory efforts to manage emotions, especially in severe cases. LSFG and LPreCG activation patterns vary with trauma age, suggesting differential effects of trauma across developmental stages. In early childhood, LSFG supports cognitive functions like working memory, flexibility, and emotional regulation (Diamond 2013; Blair and Raver 2015), which are essential for problem‐solving and adapting to new situations (Zelazo and Carlson 2012; Wessing et al. 2015). Cognitive reassessment training could support emotional regulation and PTSD recovery in children. In adulthood, LPreCG activation supports motor planning and complex tasks, including goal‐directed behavior (Bates and Goldman‐Rakic 1993; Dum and Strick 1991; Matsuzaka et al. 1992; Luppino et al. 1993). Training in complex motor skills and higher‐order cognitive functions may aid in emotional regulation recovery in adulthood PTSD patients (Grafton 2010; Hikosaka et al. 2002; Russo and Bruce 2000).

Our study found significant differences in BA48 activation across age groups, with distinct patterns in the left inferior frontal operculum, rolandic operculum, right caudate (RC), and right putamen (RPu). Greater activation in the left inferior frontal gyrus (LIFO and LIFG), regions involved in language processing and executive control (Friederici 2011; Hampshire et al. 2010), was observed in the childhood trauma group compared to adolescents. This heightened activation reflects a higher cognitive demand for emotional regulation in childhood trauma patients. The adulthood trauma group showed greater activation in LROL (BA48), linked to interoceptive awareness, indicating more severe emotional regulation difficulties and higher PTSD severity compared to adolescents. Distinct activation patterns in the right caudate (RC) and right putamen (RPu) reflect their developmental roles. RC, linked to cognitive control and memory, is vital in childhood, while the putamen supports motor control and habit maintenance, becoming more refined in adulthood (Casey et al. 2000; Alexander and Crutcher 1990; Schultz 2000). During attention‐shifting tasks like the SEAT paradigm, increased BA48 activation in PTSD suggests a higher cognitive load, reflecting difficulties in emotion regulation and attention redirection (Kanske et al. 2011). Trauma experienced at different developmental stages affects the caudate and putamen, influencing emotional development and PTSD complexity.

Childhood trauma can lead to more severe PTSD symptoms due to several factors: (1) developmental impact during a critical period of brain development, particularly in areas responsible for emotional regulation and stress response. Trauma can disrupt neural circuit formation, leading to long‐term changes in these systems (Teicher and Samson 2016). (2) Disruption of attachment and relationships, hindering resilience (Bowlby 1980), (3) maladaptive coping strategies like avoidance (Felitti et al. 1998), (4) social and environmental stressors, including ongoing stressors that further complicate recovery (McLaughlin et al. 2010), and (5) the longer the time elapsed since the trauma, the more severe the PTSD symptoms may become (Teicher and Samson 2016). Protecting childhood development through stable caregiving, quality education, and supportive communities can mitigate these effects, promoting individual well‐being and fostering healthier societies (Masten and Narayan 2012). People who experience trauma in adulthood may face emotional regulation challenges due to (1) the immediate impact of acute events overwhelming coping strategies (Bryant 2011; Ehring and Quack 2010), (2) barriers to seeking support (Brewin et al. 2000), (3) increased responsibilities raising stress levels (Bryant 2011), and (4) the recent occurrence of trauma in adults may mean that symptoms have not yet had sufficient time to subside or be mitigated through natural recovery processes or interventions (Bryant 2011).

The LpACC (BA32) in the anterior cingulate cortex(ACC) is vital for emotional processing and higher‐level cognitive functions (Fynes‐Clinton et al. 2022; Lamke et al. 2014). In PTSD, increased LpACC activation may indicate heightened emotional regulation efforts (Kaiser et al. 2018). This study found greater LpACC activation in childhood trauma survivors during emotion regulation tasks. Although there were no significant differences in PCL‐5, GAD‐7, and PHQ‐9 scores (see Table S6), several factors may explain this: (1) Self‐reports are subjective, while fMRI provides objective data; (2) PCL‐5 may not fully capture implicit emotional responses; (3) The recency of adulthood trauma could heighten perceived symptoms; (4) Childhood trauma's long‐term brain impact may require more regulation resources.

The timing of trauma exposure, particularly during developmental stages, significantly affects the brain's emotional regulation systems. Trauma during critical periods of brain development in childhood can cause lasting changes in neural circuits responsible for stress response and emotion regulation. Conversely, trauma experienced in adulthood may compromise established emotional regulation mechanisms, complicating effective coping and recovery. Understanding these age‐dependent neural differences is crucial for developing targeted therapeutic interventions. For those who have experienced childhood trauma, interventions should focus on establishing secure attachments, promoting healthy relationships, and enhancing coping strategies. For adults, interventions should prioritize alleviating immediate emotional distress, facilitating access to social support and professional resources, and managing responsibilities critical to recovery.

4.4. Limitations

The concept of trauma age, or physiological age, is a central focus of this study. It is widely accepted that more recent traumas are associated with more severe PTSD symptoms (Kessler et al. 2005; McFarlane 2010). However, our findings indicate that childhood trauma results in more severe PTSD than trauma occurring in adolescence, highlighting the importance of considering the timing of trauma in PTSD research and therapeutic strategies. However, several limitations warrant consideration when interpreting these findings and should guide future research: First, reliance on retrospective reporting of trauma history is subject to recall bias, which may limit the accuracy of reported trauma exposure, especially for early‐life trauma (Baldwin et al. 2019). Second, trauma age is inherently correlated with the time elapsed since exposure, creating ambiguity in interpreting our findings. For example, adults with childhood trauma showed higher PTSD symptom severity than those with adolescent trauma. This could be due to both the earlier developmental stage and the longer time since the trauma (Teicher and Samson 2016). Third, our subgroup analyses, stratified by trauma age, were limited by small sample sizes. This not only reduced statistical power, potentially masking significant findings and weakening our conclusions, but also constrained the robustness of normality testing and homogeneity of variance assessment, thereby affecting the validity of our statistical inferences. Finally, the lack of FDR‐surviving exploratory correlations (despite uncorrected trends in Table S5) reflects the reduced power for multidimensional brain–behavior analyses in our cohort. Future research should address these methodological constraints by incorporating prospective assessments of trauma exposure, leveraging larger sample sizes, and utilizing longitudinal designs to elucidate the complex interplay between trauma timing and PTSD outcomes, ultimately informing more targeted and effective interventions.

5. Conclusions

Our study elucidates differential activation patterns in brain regions crucial for emotion regulation by attention shifting, such as the bilateral BA48, left thalamus, frontal gyrus regions, and pregenual anterior cingulate cortex. These variations are evident in individuals with PTSD who have experienced trauma at different life stages, advancing our understanding of PTSD. This investigation reveals that the neural correlates of emotion regulation are affected by both age groups and the developmental stage at the time of trauma.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Appendix S1: Supplementary Information.

Liu, S. , Guo Y., Liu W., et al. 2025. “Trauma Timing and Its Impact on Brain Activation During Flexible Emotion Regulation in PTSD: Insights From Functional MRI .” Human Brain Mapping 46, no. 14: e70346. 10.1002/hbm.70346.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.