Hostname: page-component-74656f9884-npj9w Total loading time: 0 Render date: 2026-11-07T22:04:39.279Z Has data issue: false hasContentIssue false
  • English
  • Français

Abnormal neural activation during negative emotion processing in anxiety disorders: a coordinate-based meta-analysis

Published online by Cambridge University Press:  06 May 2024

Rui Wang Open the ORCID record for Rui Wang [Opens in a new window] ,
Lu Lu ,
Yingxue Gao ,
Hui Qiu ,
Haoran Xu ,
Xiaoqi Huang ,
John A. Sweeney ,
Qiyong Gong Open the ORCID record for Qiyong Gong [Opens in a new window]  and
Jeffrey R. Strawn
Rui Wang
Affiliation:
Department of Radiology, Huaxi MR Research Center (HMRRC), Institute of Radiology and Medical Imaging, West China Hospital of Sichuan University, Chengdu, China Research Unit of Psychoradiology, Chinese Academy of Medical Sciences, Chengdu, China
Lu Lu
Affiliation:
Department of Radiology, Huaxi MR Research Center (HMRRC), Institute of Radiology and Medical Imaging, West China Hospital of Sichuan University, Chengdu, China Psychoradiology Key Laboratory of Sichuan Province, West China Hospital of Sichuan University, Chengdu, China
Yingxue Gao
Affiliation:
Department of Radiology, Huaxi MR Research Center (HMRRC), Institute of Radiology and Medical Imaging, West China Hospital of Sichuan University, Chengdu, China Research Unit of Psychoradiology, Chinese Academy of Medical Sciences, Chengdu, China
Hui Qiu
Affiliation:
Department of Radiology, Huaxi MR Research Center (HMRRC), Institute of Radiology and Medical Imaging, West China Hospital of Sichuan University, Chengdu, China Research Unit of Psychoradiology, Chinese Academy of Medical Sciences, Chengdu, China
Haoran Xu
Affiliation:
Department of Radiology, Huaxi MR Research Center (HMRRC), Institute of Radiology and Medical Imaging, West China Hospital of Sichuan University, Chengdu, China Psychoradiology Key Laboratory of Sichuan Province, West China Hospital of Sichuan University, Chengdu, China
Xiaoqi Huang
Affiliation:
Department of Radiology, Huaxi MR Research Center (HMRRC), Institute of Radiology and Medical Imaging, West China Hospital of Sichuan University, Chengdu, China Research Unit of Psychoradiology, Chinese Academy of Medical Sciences, Chengdu, China
John A. Sweeney
Affiliation:
Department of Radiology, Huaxi MR Research Center (HMRRC), Institute of Radiology and Medical Imaging, West China Hospital of Sichuan University, Chengdu, China Department of Psychiatry and Behavioral Neuroscience, University of Cincinnati, Cincinnati, OH, USA
Qiyong Gong*
Affiliation:
Department of Radiology, Huaxi MR Research Center (HMRRC), Institute of Radiology and Medical Imaging, West China Hospital of Sichuan University, Chengdu, China Xiamen Key Lab of Psychoradiology and Neuromodulation, Department of Radiology, West China Xiamen Hospital of Sichuan University, Xiamen, China
Jeffrey R. Strawn
Affiliation:
Department of Psychiatry and Behavioral Neuroscience, University of Cincinnati, Cincinnati, OH, USA Divisions of Psychiatry and Clinical Pharmacology, Department of Pediatrics, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA
*
Corresponding author: Qiyong Gong; Email: qiyonggong@hmrrc.org.cn
Rights & Permissions [Opens in a new window]

Abstract

Background

Anxiety disorders are associated with aberrant neural responses to negative emotions. Yet, the diverse range of contrasts and stimuli used to investigate these responses has produced variable, complex, and sometimes conflicting results.

Methods

To characterize brain activation during negative emotion processing in anxiety disorders, we conducted a meta-analysis of studies contrasting activation to negative stimuli versus perceptually similar neutral stimuli and examined the differential effects of two types of visual stimuli—scenes and faces. The relevant functional magnetic resonance imaging (fMRI) studies that employed these contrasts were identified using PubMed, Web of Science, and EMBASE databases, and the meta-analysis was conducted in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

Results

Across 20 studies, patients with anxiety disorders (n = 348) had increased activation in core cortical regions of default mode and frontal–parietal networks during negative emotion processing compared to healthy controls (n = 335). Further, differential and greater regional activation was found during the processing of negative scenes than faces and greater activation was associated with sex and age of patients across studies.

Conclusions

These results highlight the importance of self-reference- and cognitive regulation-related functional disturbances in the cortex rather than emotional response-related subcortical alteration during negative emotion processing and indicate a more robust effect from emotional scenes in anxiety disorders.

Keywords

Anxiety disorderfunctional MRIemotion processingtaskmeta-analysis

Information

Type
Original Research
Information
CNS Spectrums , Volume 30 , Issue 1 , 2025 , e64
Copyright
© The Author(s), 2025. Published by Cambridge University Press

Introduction

Anxiety disorders are common psychiatric disorders characterized by impaired processing of negative emotions.1, Reference McKay and Tolin2 Functional brain imaging studies have demonstrated abnormal activation toward stimuli eliciting negative emotions in multiple neocortical and subcortical regions in patients with anxiety disorders.Reference Frick, Howner, Fischer, Kristiansson and Furmark3–Reference Padgaonkar, Phuong Uy, DePasque, Galvan and Peris6 However, these studies are frequently limited by small sample sizes, sample heterogeneity, differences in prior treatments, age of patients, and differences in the tasks used, which contribute to inconsistent findings. In this context, meta-analyses can integrate individual studies and enhance our understanding of emotional processing in anxiety disorders—and how it differs from healthy individuals. This synthesis may identify neurofunctional risk and resiliency biomarkers and elucidate the neural basis for cognitive aspects of anxiety (eg, threat perception, fear generalization). Understanding emotion processing in anxiety disorders is critical to developing therapeutic targets in these highly prevalent conditions with variable responses to traditional treatment. Indeed, there has been increased enthusiasm for translational treatment studies incorporating target engagement.Reference Bruckl, Spoormaker and Samann7, Reference Williams, Goldstein-Piekarski and Chowdhry8 Yet, these targets (ie, emotional processing circuitry) are poorly understood in anxiety disorders. Its study is mired by the heterogeneity of prior attempts to understand these targets.Reference Strawn and Levine9

Several earlier meta-analyses reveal that patients with anxiety disorders have abnormal activation during emotion processing in widespread brain regions, including the amygdala, fusiform gyrus, prefrontal cortex (PFC), anterior cingulate cortex (ACC), insula, superior temporal gyrus (STG), parietal cortex, occipital regions, and cerebellum.Reference Bruhl, Delsignore, Komossa and Weidt10–Reference Yu, Ruan and Zhang15 However, these meta-analyses differ in inclusion criteria, especially for types of task contrasts, which produce discrepancies. Specifically, they include contrasts of negative stimuli versus baseline (eg, resting state, fixation cross); negative stimuli versus neutral stimuli; negative stimuli versus positive stimuli; and negative, positive, and neutral stimuli versus baseline. Such methodological differences can meaningfully impact not only the neurofunctional processes probed by the tasks but also the magnitude of the apparent effect and the actual disorder-specific effects. Not surprisingly, prior studies comparing negative emotion stimuli to a central fixation cross instead of neutral stimuli with similar perceptual demands yield different effects. For example, Padgaonkar et al. found that aversive scenes, compared to fixation cross, increased activation in the middle frontal gyrus and precentral gyrus in youth with anxiety disorders, but comparing aversive scenes with neutral scenes was not associated with differences in activation.Reference Padgaonkar, Phuong Uy, DePasque, Galvan and Peris6 This likely relates to the involvement of regions that subserve visual and cognitive processes being recruited in processing negative faces/scenes and fixation crosses differently. Further, some studies suggest distinct brain responses to negative and positive emotional stimuli in both psychiatrically ill and healthy individuals.Reference Fusar-Poli, Placentino and Carletti16, Reference Zhang, Huang and Li17 These findings indicate that comparing negative stimuli to positive stimuli is not equivalent to comparing negative stimuli to neutral stimuli. Combining results from previous meta-analyses, which included studies with significantly different contrasts, may have resulted in a less clear and possibly less meaningful summary of previously published findings.

In addition, including studies contrasting brain responses to different types of visual stimuli (eg, negative faces versus neutral scenes, negative scenes versus neutral faces, and negative faces/scenes versus geometric shapes) that require different perceptional demands in a meta-analysis makes it harder to understand the brain activation during negative emotional processing. Goldin and colleagues found that adults with social anxiety disorder (SAD) compared to controls had altered activation in 14 clusters when comparing the faces expressing negative affect to neutral scenes, but no differences when negative and neutral scenes were compared.Reference Goldin, Manber, Hakimi, Canli and Gross18 The results of negative emotion-related brain activation become more discrepant when comparing negative emotion stimuli to geometric shapes. For example, Gaebler et al. found no differences in activation between patients with SAD and healthy controls when comparing negative faces to geometric shapes.Reference Gaebler, Daniels, Lamke, Fydrich and Walter19 However, among the same group of participants, individuals with SAD exhibited increased activation in the STG when assessing negative scenes in comparison with neutral scenes.Reference Gaebler, Daniels, Lamke, Fydrich and Walter20 Moreover, activation in dorsomedial PFC (dmPFC) was increased in patients with SAD when comparing angry faces to neutral facesReference Evans, Wright, Wedig, Gold, Pollack and Rauch21 but decreased when comparing angry faces to circular images.Reference Binelli, Muniz and Subira22 These conflicting findings highlight the substantial impact of stimulus type being used in the contrasts. Thus, to prevent potential non-emotion-related confounding effects, it is crucial to ensure a high similarity between the task stimuli being compared, with the only difference being the affective valence of the stimuli.

Moreover, in healthy individuals, faces and scenes—the two most commonly used stimuli—recruit separate neural networks during emotion processing.Reference Britton, Taylor, Sudheimer and Liberzon23, Reference Sabatinelli, Fortune and Li24 Consistent with this, different or even opposite activation patterns can be seen in response to emotional faces and scenes in patients with anxiety disorders. For instance, patients with anxiety disorders have increased ventromedial PFC activation to negative scenes compared to neutral scenesReference Strawn, Bitter and Weber25 but decreased ventromedial PFC activation to negative faces relative to neutral faces.Reference Carlisi, Hilbert, Guyer and Ernst26 Given this conflicting evidence, it is necessary to clarify negative scene- and face-specific activation patterns in anxiety disorders separately and quantitatively compare the activation pattern differences between negative scenes (versus neutral scenes) and negative faces (versus neutral faces) meta-analytically.

With these considerations in mind, we examined abnormal regional brain activation during negative emotion processing in patients with anxiety disorders by performing a meta-analysis of studies contrasting negative and neutral emotion stimuli and using stimuli that differ only in emotional valence between the two conditions. Further, we restricted this meta-analysis to generalized anxiety disorder (GAD), SAD, and separation anxiety disorder because they are highly comorbid and more characterized by anxiety symptoms than other anxiety disorders that are more characterized by episodic and discrete fear (eg, panic disorder and specific phobia).Reference Duval, Javanbakht and Liberzon27 We aimed to (1) characterize regional brain activation abnormalities during the processing of negative emotion in these anxiety disorders, (2) examine stimulus-specific activation differences (eg, faces and scenes), and (3) explore demographic features related to alterations in activation. Such an exploration could enhance our understanding of the neural mechanisms of negative emotion processing in anxiety disorders and provide insight into the neurobiological differences between processing negative emotions related to faces and scenes.

Methods

Search strategy and selection criteria

We conducted a systematic search in the PubMed, Web of Science, and EMBASE databases to retrieve relevant studies published prior to September 5, 2022, with the following search terms: (“social anxiety disorder” OR “social phobia” OR “GAD” OR “generalized anxiety disorder” OR “separation anxiety disorder” OR “anxiety”) AND (“fMRI” OR “MRI” OR “functional magnetic resonance imaging” OR “functional” OR “magnetic resonance imaging” OR “task”) AND (“emotion” OR “emotional” OR “emotion regulation” OR “emotional regulation” OR “reappraisal”) AND (“IAPS” OR “picture” OR “faces” OR “words” OR “scripts”). Additional publications were identified by manual search in the reference lists of included studies and previous meta-analyses.

Original task-based functional magnetic resonance imaging (fMRI) studies were included if they (1) were published in a peer-reviewed journal in English; (2) included patients with a primary diagnosis of GAD, SAD, or separation anxiety disorders based on the Diagnostic and Statistical Manual or International Classification of Disease criteria (no restrictions on edition used) and controls; (3) compared negative emotion stimuli with neutral emotion stimuli in the same type (ie, negative faces versus neutral faces, negative scenes versus neutral scenes); (4) directly compared brain activation between patients and controls at the whole-brain level; and (5) reported coordinates of group comparisons in stereotactic space (ie, Montreal Neurological Institute or Talairach). For studies using overlapping samples, we only recruited the one with the largest sample. Studies of patients with neurological disorders (eg, Parkinson’s disorders) with comorbid anxiety were excluded. Studies employing contrasts of negative stimuli versus baseline; negative and positive stimuli versus baseline; negative, positive, and neutral stimuli together versus baseline; negative stimuli versus positive stimuli; negative and positive stimuli versus neutral stimuli; negative scenes versus neutral faces; negative faces versus neutral scenes; negative scenes versus geometric shapes; or negative faces versus geometric shapes were excluded because of unmatched perceptual characteristics.

Two authors (R.W. and L.L.) independently conducted the literature search and selection to ensure comprehensiveness, with discrepancies resolved via discussions. The meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.Reference Moher, Shamseer and Clarke28 The protocol was registered on the International Prospective Register of Systemic Reviews (PROSPERO) (CRD42022358749).

Voxel-wise meta-analysis

This meta-analysis used the anisotropic effect size version of the signed differential mapping (AES-SDM) software package (version 5.15; http://www.sdmproject.com/software), a statistical tool for meta-analyzing neuroimaging studies.Reference Radua and Mataix-Cols29 Details of the seed-based d mapping (SDM) method have been described elsewhere.Reference Radua, Mataix-Cols and Phillips30, Reference Radua, van den Heuvel, Surguladze and Mataix-Cols31 Briefly, the peak coordinates and corresponding statistical values (ie, t, z, or p values) from each study were extracted and used to recreate effect-size maps of activation differences between the patients with anxiety disorders and controls. Both positive and negative coordinates were reconstructed in the same map so that both positive and negative effects in regions were considered simultaneously.Reference Radua and Mataix-Cols32 Finally, the mean map was obtained by combining individual maps from each study using meta-analytic calculations, weighted by intra-study variance and inter-study heterogeneity.

Statistical significance was determined using standard permutation tests with an uncorrected p < 0.005 and |Z| > 1 as the main threshold. This default threshold has been found to balance sensitivity and specificity optimally and to be an approximate equivalent to a corrected p < 0.05 in AES-SDM.Reference Radua, Mataix-Cols and Phillips30 Besides, a default AES-SDM threshold of cluster size >10 voxels was used to report significant findings.Reference Radua, Mataix-Cols and Phillips30 We also validated our primary meta-analytic findings of activation abnormalities in anxiety disorders using the latest Seed-based d Mapping with Permutation of Subject Images (SDM-PSI) toolbox (version 6.22).Reference Albajes-Eizagirre, Solanes, Vieta and Radua33 More details about the application of the SDM-PSI toolbox are available in the Supplementary materials.

Subgroup analysis and group comparisons

We conducted scene-specific and face-specific subgroup analyses separately to evaluate stimulus-specific alterations in functional activation between patients and controls. Then, we quantitatively compared altered neural activations between the two stimulus types (ie, scene-specific and face-specific) in each voxel to examine the brain response differences between these two common stimulus types. The statistical significance of task effect differences was determined using a standard randomization test.Reference Radua, van den Heuvel, Surguladze and Mataix-Cols31 To control for differences in primary diagnosis and participant age, we further performed SAD and adult subgroup meta-analyses by only including studies of patients with SAD as the primary diagnosis and studies of adults with anxiety disorders separately. Notably, due to insufficient eligible studies, we were unable to perform the GAD ( $ \kappa =6 $ ), adolescent ( $ \kappa =4 $ ), and geriatric ( $ \kappa =1 $ ) subgroup analyses and thus were unable to examine age- or disorder-specific activation differences.

Meta-regression analysis

A linear meta-regression was conducted to investigate the potential effects of relevant demographic and clinical variables. Independent variables explored by regression were the mean age and percentage of female patients. The impact of anxiety severity, the percentage of medication-naïve patients, and the status of specific psychiatric comorbidities could not be analyzed due to insufficient or heterogeneous data. To minimize the detection of spurious relationships, a more conservative threshold of uncorrected p < 0.0005 was used in the meta-regression analysis.Reference Radua and Mataix-Cols32, Reference Radua, Rubia, Canales-Rodriguez, Pomarol-Clotet, Fusar-Poli and Mataix-Cols34

Jackknife sensitivity, heterogeneity, and publication bias analysis

To test the robustness and replicability of results from the primary and subgroup analyses, a systematic whole-brain jackknife sensitivity analysis was performed by repeating the analysis n times (n = number of datasets in the meta-analysis) but discarding a different study each time. If a previously significant finding remained significant in at least 80% of the combinations of studies,Reference Wang, Gao and Tang35 it could be regarded as replicable.Reference Radua and Mataix-Cols32 Heterogeneity between studies was examined using the Q statistic.Reference Radua, Rubia, Canales-Rodriguez, Pomarol-Clotet, Fusar-Poli and Mataix-Cols34 Sources of heterogeneity were further explored through meta-regression and subgroup analyses. Funnel plots and Egger’s test in AES-SDM were used to visualize and quantitatively test the possibility of publication bias for each cluster, respectively.Reference Egger, Smith and Phillips36 The results showing p < 0.05 in Egger’s test were judged to have significant publication bias.Reference Radua, Grau and van den Heuvel37

Results

Included studies and sample characteristics

Our search generated 6740 unique records (6737 from the databases and 3 from reference lists of previous publications). Twenty studies employing negative versus neutral condition contrasts and using the same type of stimuli in negative and neutral conditions were included in the meta-analysis (Figure 1). The 20 included studies evaluated 348 individuals with anxiety disorders and 335 controls and primarily evaluated adults ( $ \kappa =15,75\% $ ). Twelve studies focused on SAD; 6 evaluated patients with GAD; and the other 2 evaluated mixed samples of patients with SAD, GAD, and separation anxiety disorder. Of note, 10 studies used negative/neutral scenes as stimuli, 8 used negative/neutral faces as stimuli, and the remaining 2 used negative/neutral words as stimuli. Demographic and clinical characteristics of participants are summarized in Table 1.

Figure 1.

Preferred reporting items for systematic reviews and meta-analyses flowchart of literature search and selection criteria.

Table 1.

Summary of the Characteristics of Studies Included in the Meta-Analysis

Abbreviations: BAI, Beck Anxiety Inventory; GAD, generalized anxiety disorder; HAMA, Hamilton Anxiety Rating Scale; LSAS, Liebowitz Social Anxiety Scale; MASC, Multidimensional Anxiety Scale for Children; MDD, major depressive disorder; OCD, obsessive–compulsive disorder; PARS, Pediatric Anxiety Rating Scale; PD, panic disorder; PSWQ, Penn State Worry Questionnaire; SAD, social anxiety disorder; SCARED, Self-Report for Childhood Anxiety Related Disorders; SepD, separation anxiety disorder; SNRI, selective noradrenalin reuptake inhibitor; SPIN, Social Phobia Inventory; SP, specific phobia; SSRI, selective serotonin reuptake inhibitor; STAI, State-Trait Anxiety Inventory.

Generalized activation alterations during negative emotion processing across the scene- and face-stimulus tasks in patients with anxiety disorders

Taking studies with contrasts of negative scenes versus neutral scenes and negative faces versus neutral faces together, compared with controls, patients with anxiety disorders had significantly generalized greater activation during negative emotion processing in the bilateral STG extending to the middle temporal gyrus (MTG), right supramarginal gyrus, bilateral posterior cingulate cortex (PCC), left dmPFC, and left cerebellum (Table 2 and Figure 2). No regions with decreased activation during negative emotion processing were observed in patients with anxiety disorders compared to controls. The subgroup analyses of patients with SAD and adults with anxiety disorders revealed similar results (Supplementary Table S1 and Supplementary Figure S1). When using SDM-PSI for validation, no significant clusters survived correction for multiple comparisons.

Table 2.

Brain Activations during Negative Emotion Processing in Patients with Anxiety Disorders Compared to Controls

Abbreviations: MNI, Montreal Neurological Institute; SDM, seed-based d mapping.

Figure 2.

Activations during negative emotion processing in patients with anxiety disorders compared to controls. The red clusters indicate increased activation (hyperactivation); the blue clusters indicate decreased activation (hypoactivation). Abbreviations: ACC, anterior cingulate cortex; B, bilateral; dlPFC, dorsolateral prefrontal cortex; dmPFC, dorsomedial prefrontal cortex; L, left; MTG, middle temporal gyrus; PCC, posterior cingulate cortex; R, right; SPG, superior parietal gyrus; STG, superior temporal gyrus; vlPFC, ventrolateral prefrontal cortex.

Differed activation alterations during negative emotion processing between scene-stimulus tasks and face-stimulus tasks in patients with anxiety disorders

In studies comparing negative scenes to neutral scenes, patients with anxiety disorders had significantly increased activation in the left STG extending to the MTG, bilateral dmPFC, bilateral PCC, right superior parietal gyrus (SPG), and right STG relative to controls. In studies comparing negative faces to neutral faces, activation was increased in the left cerebellum and right ventrolateral PFC and decreased in the bilateral ACC, right dorsolateral PFC, left SPG, left PCC, left precentral gyrus, right thalamus, and right supramarginal gyrus (Table 2 and Figure 2).

In the comparison of face-stimulus studies and scene-stimulus studies, the latter produced greater activation in bilateral ACC extending to dmPFC, right SPG, bilateral PCC, and left MTG, as well as decreased activation in the left cerebellum in patients with anxiety disorders (Table 2 and Figure 2).

Meta-regression

In meta-regression analyses, older patients with anxiety disorders had greater activation in the right MTG (r = 0.657, p < 0.001) and supramarginal gyrus (r = 0.653, p < 0.001). Additionally, studies with more female patients had greater activation in left STG (r = 0.330, p < 0.001) and PCC (r = 0.322, p < 0.001) (Figure 3).

Figure 3.

The results of meta-regression analysis. Abbreviations: MTG, middle temporal gyrus; PCC, posterior cingulate cortex; STG, superior temporal gyrus.

Jackknife sensitivity, heterogeneity, and publication bias analysis

For the results of the main analysis, sensitivity analyses revealed that the clusters in bilateral STG extending to MTG were highly replicable, with all 20 combinations of studies preserved. The right supramarginal gyrus and left dmPFC remained significant in all but one combination, while the remaining clusters in bilateral PCC and cerebellum remained significant in all but 3 or 4 combinations. Most results in subgroup analyses were robust enough to survive more than 80% of iterations under jackknife sensitivity analyses (Supplementary Tables S2–S6).

Only bilateral PCC and the left cerebellum among significant clusters identified in our primary analysis showed considerable heterogeneity between studies. Egger’s tests were nonsignificant, and funnel plots (Supplementary Figure S2) suggested a low likelihood of publication bias.

Discussion

Our meta-analysis is unique in that it focuses on task-based fMRI studies employing the same stimulus type in both negative and neutral emotion stimuli and direct contrasts of negative and neutral stimuli. This approach minimizes confounding effects arising from experimental paradigms, a significant limitation of prior meta-analyses. It reveals distinct patterns of brain activation during negative emotion processing in patients with anxiety disorders. Herein, we found that, compared with controls, patients with anxiety disorders had increased activation in the STG, MTG, supramarginal gyrus, PCC, dmPFC, and cerebellum. These findings were robust in subgroup analyses, including studies with patients with a primary diagnosis of SAD only, adults with anxiety disorders only, and negative and neutral scenes as stimuli only. Our meta-analysis excluded studies with unmatched contrast and stimulus types, such as using resting-state fMRI or positive emotion as the comparator for stimuli with a negative emotional valence, thus improving the utility of study findings for understanding the alterations in brain processing of negative emotion in patients with anxiety disorders.

The dmPFC, PCC, STG, and MTG belong to the default mode network (DMN),Reference Koster, Crombez, Verschuere, Van Damme and Wiersema49 which subserves self-referential processing (eg, “Is this relevant to me?” or “Can I handle this?”) predominantly negative self-referential processing (eg, “I am ugly” and “you are boring to me”).Reference Ford, Grasso, Elhai, Courtois, Ford, Grasso, Elhai and Courtois50, Reference Moran, Heatherton and Kelley51 Accumulating evidence indicates that self-referential processes involve increased DMN activation, especially during exposure to negative stimuli in patients with anxiety disorders.Reference Blair, Geraci and Otero52, Reference Yoon, Seo, Kim and Choo53 DMN hyperactivation in patients with anxiety disorders suggests accentuated self-referential thinking in the presence of negative stimuli, which might relate to negative interpretation biases (ie, interpreting negative, neutral, and other emotionally ambiguous stimuli more negative)—a hallmark of these disorders.Reference Chen, Short and Kemps54 SPG and supramarginal gyrus, important frontal–parietal network (FPN) regions, are associated with emotion and cognitive regulation. Increased FPN activation may suggest that patients with anxiety disorders needed more regulation and involvement to handle the negative stimuli compared with healthy controls. In addition to the increased DMN and FPN activation, we found here, prior studies reveal that patients with anxiety disorders have decreased functional connectivity between DMN and FPN during restReference Xu, Van Dam and Feng55 and emotion processing.Reference Strawn, Bitter and Weber25 Increased FPN activation might be a compensatory alteration of the impaired FPN–DMN connectivity that facilitates regulation of a patient’s response to emotional stimuli.

In this meta-analysis, negative scenes and faces evoked distinct brain networks, consistent with findings from healthy individuals.Reference Sabatinelli, Fortune and Li24 In contrasts of negative scenes and negative faces, we observed greater robustness of case–control differences during processing negative scenes compared to negative faces. Negative scenes evoked greater activation in ACC, dmPFC, PCC, SPG, and MTG (ie, DMN and FPN) but lower activation in the cerebellum relative to the negative faces. There are several potential reasons for this difference that may relate to psychological aspects of anxiety disorders. First, negative scenes may induce greater emotional arousal than negative facesReference Britton, Taylor, Sudheimer and Liberzon23 and thus produce greater self-referential processing and increased DMN activation. Second, the increased complexity of negative scenes may require greater perception and regulation demands on the FPN,Reference Adhikari, Lerner and Finkelstein56–Reference Pico-Perez, Radua, Steward, Menchon and Soriano-Mas59 which may contribute to the greater activation we observed in SPG when processing negative scenes relative to negative faces. Third, when viewing negative scene stimuli, patients with anxiety disorders may allocate more top-down attentional resources for threat scanning,Reference Adhikari, Lerner and Finkelstein56, Reference Hariri, Tessitore, Mattay, Fera and Weinberger58 which might interpret why negative scenes induce greater activation in the FPN. The negative face-related cerebellar hyperactivation is consistent with the study by Reisch et al. in which cerebellar activation was observed when individuals viewed negative faces but not when they viewed negative scenes or words.Reference Reisch, Wegrzyn, Woermann, Bien and Kissler60 This direct condition comparison is in line with the evolving consensus that implicates the cerebellum in perceiving, encoding, and discriminating facial expressions,Reference Fusar-Poli, Placentino and Carletti16, Reference Adamaszek, D’Agata and Kirkby61–Reference Turner, Paradiso and Marvel63 especially negative emotions.Reference Adamaszek, D’Agata and Ferrucci64, Reference Northoff and Bermpohl65 Moreover, the face-specific increased activation in the cerebellum as well as the opposite activation pattern in PCC in the face- and scene-specific subgroup analyses also explained the heterogeneity we found in these two regions, indicating different neural responses of emotion processing between scenes and faces although they both belonged to the visual modality. Taken together, our results suggested that the type of stimuli influences the robustness of alterations in brain responses to negative emotions in patients with anxiety disorders. While it is difficult to compare the intensity and specific aspects of emotion elicited by the facial and scene stimuli used in prior studies, our findings suggest that scene stimuli may be more useful for studying alterations in brain systems supporting emotional responses in patients with anxiety disorders, which could be important in selecting stimuli for genetic association studies or studies of treatment response.

The meta-regression analyses revealed that the mean age and proportion of female participants were positively correlated with the degree of brain activation. Specifically, studies with older patients had greater activation in the right MTG and supramarginal gyrus during negative emotion processing. In contrast, studies with more women had more activation in the left STG and PCC. This age effect on brain activation in response to negative emotions, together with a prior study, showed that older patients with GAD had stronger resting-state functional connectivity within DMN,Reference Andreescu, Sheu, Tudorascu, Walker and Aizenstein66 which raises the possibility that anxiety disorders might exacerbate the effect of age on brain functional activation and connectivity especially within the DMN. The relationship between STG activation and the proportion of female patients replicates an earlier study of healthy adults suggesting that unpleasant stimuli induce greater STG activation in women compared to men,Reference Hofer, Siedentopf and Ischebeck67 which could relate to the clinical observations of more severe or persistent symptoms in women.Reference Hallers-Haalboom, Maas, Kunst and Bekker68

Compared to previous meta-analysesReference Bruhl, Delsignore, Komossa and Weidt10–Reference Yu, Ruan and Zhang15 that allowed for significant contrasts and stimulus heterogeneity, we did not observe differences in brain activation while processing negative emotion in the amygdala, insula, fusiform gyrus, or occipital cortex. The fusiform gyrus contributes to face perception,Reference Haxby, Hoffman and Gobbini69–Reference Puce, Allison, Asgari, Gore and McCarthy71 and its activation in previous meta-analyses could thus in part be related to the inclusion of negative faces to resting-state conditions or geometric figures rather than neutral faces. Studies comparing complex negative scenes with simple perceptual conditions or rest may produce greater occipital activation related to visual perception. Unexpectedly, our analysis did not show increased activation in the amygdala and insula, the key regions of the limbic network that are associated with emotional response, in anxiety disorders. Instead, our findings suggest that alterations in neocortical areas that provide self-referential and regulatory control of these emotional response regions may be more impaired than regions that are more primary to emotional responses. Several factors may account for the lack of group differences in amygdala and insula activation. First, the initial emotional responses to task stimuli may be robust in both patients and controls, which could limit the ability to detect differences in these deeper limbic regions. Consistent with this speculation, Dixon et al. found that patients with SAD and healthy controls had similar brain activation during emotion processing and did not observe group differences.Reference Dixon, Moodie and Goldin72 Parametric studies regulating the degree of emotionality of task stimuli might help address these issues. Second, the small sample sizes, inadequate correction for multiple comparisons, and region of interest analyses focused on the amygdala in previous studies may artificially boost the amygdala’s significance of this structure’s role in the pathoetiology of anxiety disorders. Consistent with this possibility, a recent meta-analysis of brain activation during emotion processing and mega-analysis examining structural alterations in anxiety disorders at the whole-brain level failed to confirm a significant difference in the amygdala.Reference Weber-Goericke and Muehlhan14, Reference Harrewijn, Cardinale and Groenewold73 Last, emotion processing is associated with distributed and interconnected brain regions, so it might be the functional connectivity between regions of emotional response—amygdala and insula—and regions of emotion regulation but not the activation in regions of emotional response per se that is impaired in anxiety disorders.

While this is the first meta-analysis of anxiety disorders to focus on studies with homogeneous task-control comparisons, there are several limitations. First, like all voxel-wise meta-analytic approaches, we used coordinates reported in original studies rather than raw t-statistic maps or individual image data, which limits the precision of activation effects and group differences.Reference Radua, van den Heuvel, Surguladze and Mataix-Cols31, Reference Radua and Mataix-Cols32 Meta-analysis with subject-level behavior and imaging data during emotion processing may provide more granular results for this field. Second, most of the included studies recruited patients with anxiety disorders and comorbid conditions (eg, major depressive disorder). While this increases the real-world applicability, it may decrease the specificity of findings. The influence of comorbidity on brain activation during negative emotion processing is unclear.Reference Beesdo, Lau and Guyer74 Third, most patients in the included studies were currently or previously treated with medication, which may alter activation profiles in response to emotion tasks. Studies of untreated patients are needed to define such possible effects. Fourth, although the leave-one-out jackknife sensitivity analysis was performed to test the robustness and replicability of findings from meta-analyses, it might fail to overcome bias from a small set of highly influential studies since it only removed one study each time and repeated the analysis. Last, unfortunately, our findings did not survive the Family Wise Error–Threshold-Free Cluster Enhancement (FWE-TFCE) correction of SDM-PSI, and we believed two factors might have caused this discrepancy. First, AES-SDM has been reported to be more voxel-based sensitive than SDM-PSI when only peak coordinates are available. Second, SDM-PSI is developed with a deep focus on the control of FWE correction, which may be associated with too conservative p values.Reference Albajes-Eizagirre, Solanes, Vieta and Radua33 Future meta-analysis, including more high-quality studies and having higher statistical power, is needed for validation.

Conclusion

We examined brain activation to negative emotions in patients with anxiety disorders using task-based fMRI studies contrasting responses to negative and neutral stimuli that employed similar, affectively neutral control stimuli. Patients with anxiety disorders (compared to healthy individuals) had increased activation in core regions of DMN and FPN in response to negative relative to neutral emotions. Further, the processing of affective scenes (compared to faces) produced differential and greater regional activation. Our findings underscore the importance of self-reference- and cognitive regulation-related functional disturbances during negative emotion processing in anxiety disorders. Taken together, these results suggest that more robust effects can be observed by using emotional scenes compared to emotional faces, and a lack of consistent activation in emotional response-related limbic regions, including amygdala and insula. Finally, our findings underscore the need to develop a better understanding of neurofunctional biomarkers using more standardized paradigms and the challenge when comparing results between studies that have used differing approaches.Reference Baumel, Lu and Huang75

Supplementary material

The supplementary material for this article can be found at http://doi.org/10.1017/S1092852924000166.

Author contribution

Data curation: R.W., L.L.; Formal analysis: R.W., H.X.; Methodology: R.W., H.Q.; Resources: R.W., J.H., L.L.; Software: R.W., Y.G.; Writing – original draft: R.W., L.L.; Supervision: J.H., L.L., J.R.S.; Conceptualization: L.L., J.A.S.; Funding acquisition: L.L., Q.G.; Writing – review & editing: L.L., J.R.S., J.A.S., Q.G.; Investigation: H.X.; Validation: H.X.; Project administration: Q.G.

Financial support

This work was supported by the National Natural Science Foundation of China (LL, grant number 82202116), the Natural Science Foundation of Sichuan Province (LL, grant number 24NSFSC7937), the National Key R&D Program of China (QG, grant number 2022YFC2009900), and the National Institute of Child Health and Development (JRS, grant number R01 HD098757).

Disclosure

Dr. Sweeney consulted with VeraSci. Dr. Strawn has received research support from AbbVie, PCORI, and the National Institutes of Health. He provided consultation to Otsuka, Intracellular Therapeutics, and Cerevel. He received royalties from Springer Publishing, Cambridge University Press, and UpToDate, as well as material support from Myriad. He has also received honoraria from the American Academy of Pediatrics, the American Academy of Child & Adolescent Psychiatry, and Medscape. The remaining authors have nothing to disclose.

Footnotes

R.W. and L.L. contributed equally.

†

Deceased.

References

Collaborators GBDMD. Global, regional, and national burden of 12 mental disorders in 204 countries and territories, 1990−2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Psychiatry. 2022;9(2):137–150.CrossRefGoogle Scholar
McKay, D, Tolin, DF. Empirically supported psychological treatments and the Research Domain Criteria (RDoC). J Affect Disord. 2017;216:78–88.10.1016/j.jad.2016.10.018CrossRefGoogle ScholarPubMed
Frick, A, Howner, K, Fischer, H, Kristiansson, M, Furmark, T. Altered fusiform connectivity during processing of fearful faces in social anxiety disorder. Transl Psychiatry. 2013;3(10):e312.CrossRefGoogle ScholarPubMed
Heitmann, CY, Feldker, K, Neumeister, P, et al. Brain activation to task-irrelevant disorder-related threat in social anxiety disorder: the impact of symptom severity. Neuroimage Clin. 2017;14:323–333.10.1016/j.nicl.2017.01.020CrossRefGoogle ScholarPubMed
Heitmann, CY, Feldker, K, Neumeister, P, et al. Abnormal brain activation and connectivity to standardized disorder-related visual scenes in social anxiety disorder. Hum. Brain Mapp. 2016;37(4):1559–1572.10.1002/hbm.23120CrossRefGoogle ScholarPubMed
Padgaonkar, NT, Phuong Uy, J, DePasque, S, Galvan, A, Peris, TS. Neural correlates of emotional reactivity and regulation in youth with and without anxiety. Depress. Anxiety. 2021;38(8):804–815.10.1002/da.23154CrossRefGoogle Scholar
Bruckl, TM, Spoormaker, VI, Samann, PG, et al. The biological classification of mental disorders (BeCOME) study: a protocol for an observational deep-phenotyping study for the identification of biological subtypes. BMC Psychiatry. 2020;20(1):213.CrossRefGoogle Scholar
Williams, LM, Goldstein-Piekarski, AN, Chowdhry, N, et al. Developing a clinical translational neuroscience taxonomy for anxiety and mood disorder: protocol for the baseline-follow up Research domain criteria Anxiety and Depression ("RAD") project. BMC Psychiatry. 2016;16:68.10.1186/s12888-016-0771-3CrossRefGoogle ScholarPubMed
Strawn, JR, Levine, A. Treatment response biomarkers in anxiety disorders: from neuroimaging to neuronally-derived extracellular vesicles and beyond. Biomark Neuropsychiatry. 2020;3:100024.10.1016/j.bionps.2020.100024CrossRefGoogle ScholarPubMed
Bruhl, AB, Delsignore, A, Komossa, K, Weidt, S. Neuroimaging in social anxiety disorder-a meta-analytic review resulting in a new neurofunctional model. Neurosci. Biobehav. Rev. 2014;47:260–280.10.1016/j.neubiorev.2014.08.003CrossRefGoogle Scholar
Etkin, A, Wager, TD. Functional neuroimaging of anxiety: a meta-analysis of emotional processing in PTSD, social anxiety disorder, and specific phobia. Am. J. Psychiatry. 2007;164(10):1476–1488.10.1176/appi.ajp.2007.07030504CrossRefGoogle ScholarPubMed
Gentili, C, Cristea, IA, Angstadt, M, et al. Beyond emotions: A meta-analysis of neural response within face processing system in social anxiety. Exp. Biol. Med. (Maywood). 2016;241(3):225–237.10.1177/1535370215603514CrossRefGoogle ScholarPubMed
Kolesar, TA, Bilevicius, E, Wilson, AD, Kornelsen, J. Systematic review and meta-analyses of neural structural and functional differences in generalized anxiety disorder and healthy controls using magnetic resonance imaging. Neuroimage. Clin. 2019;24:102016.10.1016/j.nicl.2019.102016CrossRefGoogle ScholarPubMed
Weber-Goericke, F, Muehlhan, M. A quantitative meta-analysis of fMRI studies investigating emotional processing in excessive worriers: application of activation likelihood estimation analysis. J. Affect. Disord. 2019;243:348–359.10.1016/j.jad.2018.09.049CrossRefGoogle ScholarPubMed
Yu, X, Ruan, Y, Zhang, Y, et al. Cognitive neural mechanism of social anxiety disorder: a meta-analysis based on fMRI Studies. Int. J. Environ. Res. Public Health. 2021;18(11):5556.10.3390/ijerph18115556CrossRefGoogle ScholarPubMed
Fusar-Poli, P, Placentino, A, Carletti, F, et al. Functional atlas of emotional faces processing: a voxel-based meta-analysis of 105 functional magnetic resonance imaging studies. J. Psychiatry Neurosci. 2009;34(6):418–432.CrossRefGoogle ScholarPubMed
Zhang, Z, Huang, P, Li, S, et al. Neural mechanisms underlying the processing of emotional stimuli in individuals with depression: an ALE meta-analysis study. Psychiatry Res. 2022;313:114598.10.1016/j.psychres.2022.114598CrossRefGoogle ScholarPubMed
Goldin, PR, Manber, T, Hakimi, S, Canli, T, Gross, JJ. Neural bases of social anxiety disorder: emotional reactivity and cognitive regulation during social and physical threat. Arch. Gen. Psychiatry. 2009;66(2):170–180.10.1001/archgenpsychiatry.2008.525CrossRefGoogle ScholarPubMed
Gaebler, M, Daniels, JK, Lamke, JP, Fydrich, T, Walter, H. Heart rate variability and its neural correlates during emotional face processing in social anxiety disorder. Biol. Psychol. 2013;94(2):319–330.10.1016/j.biopsycho.2013.06.009CrossRefGoogle ScholarPubMed
Gaebler, M, Daniels, JK, Lamke, JP, Fydrich, T, Walter, H. Behavioural and neural correlates of self-focused emotion regulation in social anxiety disorder. J. Psychiatry Neurosci. 2014;39(4):249–258.CrossRefGoogle ScholarPubMed
Evans, KC, Wright, CI, Wedig, MM, Gold, AL, Pollack, MH, Rauch, SL. A functional MRI study of amygdala responses to angry schematic faces in social anxiety disorder. Depress. Anxiety. 2008;25(6):496–505.10.1002/da.20347CrossRefGoogle ScholarPubMed
Binelli, C, Muniz, A, Subira, S, et al. Facial emotion processing in patients with social anxiety disorder and Williams-Beuren syndrome: an fMRI study. J. Psychiatry Neurosci. 2016;41(3):182–191.10.1503/jpn.140384CrossRefGoogle ScholarPubMed
Britton, JC, Taylor, SF, Sudheimer, KD, Liberzon, I. Facial expressions and complex IAPS pictures: common and differential networks. Neuroimage. 2006;31(2):906–919.10.1016/j.neuroimage.2005.12.050CrossRefGoogle ScholarPubMed
Sabatinelli, D, Fortune, EE, Li, Q, et al. Emotional perception: meta-analyses of face and natural scene processing. Neuroimage. 2011;54(3):2524–2533.10.1016/j.neuroimage.2010.10.011CrossRefGoogle ScholarPubMed
Strawn, JR, Bitter, SM, Weber, WA, et al. Neurocircuitry of generalized anxiety disorder in adolescents: a pilot functional neuroimaging and functional connectivity study. Depress. Anxiety. 2012;29(11):939–947.10.1002/da.21961CrossRefGoogle ScholarPubMed
Carlisi, CO, Hilbert, K, Guyer, AE, Ernst, M. Sleep-amount differentially affects fear-processing neural circuitry in pediatric anxiety: a preliminary fMRI investigation. Cogn. Affect. Behav. Neurosci. 2017;17(6):1098–1113.10.3758/s13415-017-0535-7CrossRefGoogle ScholarPubMed
Duval, ER, Javanbakht, A, Liberzon, I. Neural circuits in anxiety and stress disorders: a focused review. Ther. Clin. Risk Manag. 2015;11:115–126.Google ScholarPubMed
Moher, D, Shamseer, L, Clarke, M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Systematic Reviews. 2015;4:1.10.1186/2046-4053-4-1CrossRefGoogle ScholarPubMed
Radua, J, Mataix-Cols, D. Meta-analytic methods for neuroimaging data explained. Biology of Mood & Anxiety Disorders. 2012;2:6.10.1186/2045-5380-2-6CrossRefGoogle ScholarPubMed
Radua, J, Mataix-Cols, D, Phillips, ML, et al. A new meta-analytic method for neuroimaging studies that combines reported peak coordinates and statistical parametric maps. Eur. Psychiatry. 2012;27(8):605–611.CrossRefGoogle ScholarPubMed
Radua, J, van den Heuvel, OA, Surguladze, S, Mataix-Cols, D. Meta-analytical comparison of voxel-based morphometry studies in obsessive-compulsive disorder vs other anxiety disorders. Arch. Gen. Psychiatry. 2010;67(7):701–711.10.1001/archgenpsychiatry.2010.70CrossRefGoogle ScholarPubMed
Radua, J, Mataix-Cols, D. Voxel-wise meta-analysis of grey matter changes in obsessive-compulsive disorder. Br. J. Psychiatry. 2009;195(5):393–402.10.1192/bjp.bp.108.055046CrossRefGoogle ScholarPubMed
Albajes-Eizagirre, A, Solanes, A, Vieta, E, Radua, J. Voxel-based meta-analysis via permutation of subject images (PSI): theory and implementation for SDM. Neuroimage. 2019;186:174–184.10.1016/j.neuroimage.2018.10.077CrossRefGoogle ScholarPubMed
Radua, J, Rubia, K, Canales-Rodriguez, EJ, Pomarol-Clotet, E, Fusar-Poli, P, Mataix-Cols, D. Anisotropic kernels for coordinate-based meta-analyses of neuroimaging studies. Front. Psychiatry. 2014;5:13.10.3389/fpsyt.2014.00013CrossRefGoogle ScholarPubMed
Wang, Y, Gao, Y, Tang, S, et al. Large-scale network dysfunction in the acute state compared to the remitted state of bipolar disorder: a meta-analysis of resting-state functional connectivity. EBioMedicine. 2020;54:102742.10.1016/j.ebiom.2020.102742CrossRefGoogle Scholar
Egger, M, Smith, GD, Phillips, AN. Meta-analysis: principles and procedures. BMJ. 1997;315(7121):1533–1537.10.1136/bmj.315.7121.1533CrossRefGoogle ScholarPubMed
Radua, J, Grau, M, van den Heuvel, OA, et al. Multimodal voxel-based meta-analysis of white matter abnormalities in obsessive-compulsive disorder. Neuropsychopharmacology. 2014;39(7):1547–1557.10.1038/npp.2014.5CrossRefGoogle ScholarPubMed
Price, RB, Eldreth, DA, Mohlman, J. Deficient prefrontal attentional control in late-life generalized anxiety disorder: an fMRI investigation. Transl Psychiatry. 2011;1(10):e46.10.1038/tp.2011.46CrossRefGoogle ScholarPubMed
Palm, ME, Elliott, R, McKie, S, Deakin, JF, Anderson, IM. Attenuated responses to emotional expressions in women with generalized anxiety disorder. Psychol. Med. 2011;41(5):1009–1018.CrossRefGoogle ScholarPubMed
Shah, SG, Klumpp, H, Angstadt, M, Nathan, PJ, Phan, KL. Amygdala and insula response to emotional images in patients with generalized social anxiety disorder. J. Psychiatry Neurosci. 2009;34(4):296–302.10.1139/jpn.0939CrossRefGoogle ScholarPubMed
Monk, CS, Telzer, EH, Mogg, K, et al. Amygdala and ventrolateral prefrontal cortex activation to masked angry faces in children and adolescents with generalized anxiety disorder. Arch. Gen. Psychiatry. 2008;65(5):568–576.10.1001/archpsyc.65.5.568CrossRefGoogle ScholarPubMed
Amir, N, Klumpp, H, Elias, J, Bedwell, JS, Yanasak, N, Miller, LS. Increased activation of the anterior cingulate cortex during processing of disgust faces in individuals with social phobia. Biol. Psychiatry. 2005;57(9):975–981.10.1016/j.biopsych.2005.01.044CrossRefGoogle ScholarPubMed
Bruhl, AB, Rufer, M, Delsignore, A, Kaffenberger, T, Jancke, L, Herwig, U. Neural correlates of altered general emotion processing in social anxiety disorder. Brain Res. 2011;1378:72–83.10.1016/j.brainres.2010.12.084CrossRefGoogle ScholarPubMed
Labuschagne, I, Phan, KL, Wood, A, et al. Medial frontal hyperactivity to sad faces in generalized social anxiety disorder and modulation by oxytocin. Int. J. Neuropsychopharmacol. 2012;15(7):883–896.10.1017/S1461145711001489CrossRefGoogle ScholarPubMed
Nakao, T, Sanematsu, H, Yoshiura, T, et al. fMRI of patients with social anxiety disorder during a social situation task. Neurosci. Res. 2011; 69(1):67–72.10.1016/j.neures.2010.09.008CrossRefGoogle ScholarPubMed
Schmidt, S, Mohr, A, Miltner, WH, Straube, T. Task-dependent neural correlates of the processing of verbal threat-related stimuli in social phobia. Biol. Psychol. 2010;84(2):304–312.10.1016/j.biopsycho.2010.03.005CrossRefGoogle ScholarPubMed
Hiser, J, Schneider, B, Koenigs, M. Uncertainty potentiates neural and cardiac responses to visual stimuli in anxiety disorders. Biol Psychiatry Cogn Neurosci Neuroimaging. 2021;6(7):725–734.Google ScholarPubMed
Li, J, Zhong, Y, Ma, Z, et al. Emotion reactivity-related brain network analysis in generalized anxiety disorder: a task fMRI study. BMC Psychiatry. 2020;20(1):429.10.1186/s12888-020-02831-6CrossRefGoogle ScholarPubMed
Koster, EH, Crombez, G, Verschuere, B, Van Damme, S, Wiersema, JR Components of attentional bias to threat in high trait anxiety: facilitated engagement, impaired disengagement, and attentional avoidance. Behav. Res. Ther. 2006;44(12):1757–1771.CrossRefGoogle ScholarPubMed
Ford, JD, Grasso, DJ, Elhai, JD, Courtois, CA. 5 - neurobiology of traumatic stress disorders and their impact on physical health. In: Ford, JD, Grasso, DJ, Elhai, JD, Courtois, CA, eds. Posttraumatic Stress Disorder (2nd ed. San Diego: Academic Press; 2015:183–232.10.1016/B978-0-12-801288-8.00005-4CrossRefGoogle Scholar
Moran, JM, Heatherton, TF, Kelley, WM. Modulation of cortical midline structures by implicit and explicit self-relevance evaluation. Soc. Neurosci. 2009;4(3):197–211.10.1080/17470910802250519CrossRefGoogle ScholarPubMed
Blair, KS, Geraci, M, Otero, M, et al. Atypical modulation of medial prefrontal cortex to self-referential comments in generalized social phobia. Psychiatry Res. 2011;193(1):38–45.10.1016/j.pscychresns.2010.12.016CrossRefGoogle ScholarPubMed
Yoon, HJ, Seo, EH, Kim, JJ, Choo, IH. Neural correlates of self-referential processing and their clinical implications in social anxiety disorder. Clin Psychopharmacol Neurosci. 2019;17(1):12–24.10.9758/cpn.2019.17.1.12CrossRefGoogle ScholarPubMed
Chen, J, Short, M, Kemps, E. Interpretation bias in social anxiety: a systematic review and meta-analysis. J. Affect. Disord. 2020;276:1119–1130.10.1016/j.jad.2020.07.121CrossRefGoogle ScholarPubMed
Xu, J, Van Dam, NT, Feng, C, et al. Anxious brain networks: a coordinate-based activation likelihood estimation meta-analysis of resting-state functional connectivity studies in anxiety. Neurosci. Biobehav. Rev. 2019;96:21–30.10.1016/j.neubiorev.2018.11.005CrossRefGoogle ScholarPubMed
Adhikari, A, Lerner, TN, Finkelstein, J, et al. Basomedial amygdala mediates top-down control of anxiety and fear. Nature. 2015;527(7577):179–185.10.1038/nature15698CrossRefGoogle ScholarPubMed
Etkin, A, Egner, T, Kalisch, R. Emotional processing in anterior cingulate and medial prefrontal cortex. Trends Cogn. Sci. 2011;15(2):85–93.10.1016/j.tics.2010.11.004CrossRefGoogle ScholarPubMed
Hariri, AR, Tessitore, A, Mattay, VS, Fera, F, Weinberger, DR. The amygdala response to emotional stimuli: a comparison of faces and scenes. Neuroimage. 2002;17(1):317–323.10.1006/nimg.2002.1179CrossRefGoogle ScholarPubMed
Pico-Perez, M, Radua, J, Steward, T, Menchon, JM, Soriano-Mas, C. Emotion regulation in mood and anxiety disorders: a meta-analysis of fMRI cognitive reappraisal studies. Prog. Neuropsychopharmacol. Biol. Psychiatry. 2017;79(Pt B):96–104.10.1016/j.pnpbp.2017.06.001CrossRefGoogle ScholarPubMed
Reisch, LM, Wegrzyn, M, Woermann, FG, Bien, CG, Kissler, J. Negative content enhances stimulus-specific cerebral activity during free viewing of pictures, faces, and words. Hum. Brain Mapp. 2020;41(15):4332–4354.10.1002/hbm.25128CrossRefGoogle ScholarPubMed
Adamaszek, M, D’Agata, F, Kirkby, KC, et al. Impairment of emotional facial expression and prosody discrimination due to ischemic cerebellar lesions. Cerebellum. 2014;13(3):338–345.10.1007/s12311-013-0537-0CrossRefGoogle ScholarPubMed
Ferrucci, R, Giannicola, G, Rosa, M, et al. Cerebellum and processing of negative facial emotions: cerebellar transcranial DC stimulation specifically enhances the emotional recognition of facial anger and sadness. Cogn. Emot. 2012;26(5):786–799.10.1080/02699931.2011.619520CrossRefGoogle ScholarPubMed
Turner, BM, Paradiso, S, Marvel, CL, et al. The cerebellum and emotional experience. Neuropsychologia. 2007;45(6):1331–1341.10.1016/j.neuropsychologia.2006.09.023CrossRefGoogle ScholarPubMed
Adamaszek, M, D’Agata, F, Ferrucci, R, et al. Consensus paper: cerebellum and emotion. Cerebellum. 2017;16(2):552–576.10.1007/s12311-016-0815-8CrossRefGoogle ScholarPubMed
Northoff, G, Bermpohl, F. Cortical midline structures and the self. Trends Cogn. Sci.. 2004;8(3):102–107.10.1016/j.tics.2004.01.004CrossRefGoogle ScholarPubMed
Andreescu, C, Sheu, LK, Tudorascu, D, Walker, S, Aizenstein, H. The ages of anxiety--differences across the lifespan in the default mode network functional connectivity in generalized anxiety disorder. Int. J. Geriatr. Psychiatry. 2014;29(7):704–712.10.1002/gps.4051CrossRefGoogle ScholarPubMed
Hofer, A, Siedentopf, CM, Ischebeck, A, et al. Gender differences in regional cerebral activity during the perception of emotion: a functional MRI study. Neuroimage. 2006;32(2):854–862.10.1016/j.neuroimage.2006.03.053CrossRefGoogle ScholarPubMed
Hallers-Haalboom, ET, Maas, J, Kunst, LE, Bekker, MHJ. The role of sex and gender in anxiety disorders: being scared “like a girl”?. Handb. Clin. Neurol. 2020;175:359–368.Google Scholar
Haxby, JV, Hoffman, EA, Gobbini, MI. Human neural systems for face recognition and social communication. Biol. Psychiatry. 2002;51(1):59–67.CrossRefGoogle ScholarPubMed
Kanwisher, N, Yovel, G. The fusiform face area: a cortical region specialized for the perception of faces. Philos. Trans. R. Soc. Lond. B Biol. Sci. 2006;361(1476):2109–2128.10.1098/rstb.2006.1934CrossRefGoogle ScholarPubMed
Puce, A, Allison, T, Asgari, M, Gore, JC, McCarthy, G. Differential sensitivity of human visual cortex to faces, letterstrings, and textures: a functional magnetic resonance imaging study. J. Neurosci. 1996;16(16):5205–5215.10.1523/JNEUROSCI.16-16-05205.1996CrossRefGoogle ScholarPubMed
Dixon, ML, Moodie, CA, Goldin, PR, et al. Frontoparietal and default mode network contributions to self-referential processing in social anxiety disorder. Cogn. Affect. Behav. Neurosci. 2022;22(1):187–198.10.3758/s13415-021-00933-6CrossRefGoogle ScholarPubMed
Harrewijn, A, Cardinale, EM, Groenewold, NA, et al. Cortical and subcortical brain structure in generalized anxiety disorder: findings from 28 research sites in the ENIGMA-Anxiety Working Group. Transl Psychiatry. 2021;11(1):502.10.1038/s41398-021-01622-1CrossRefGoogle ScholarPubMed
Beesdo, K, Lau, JY, Guyer, AE, et al. Common and distinct amygdala-function perturbations in depressed vs anxious adolescents. Arch. Gen. Psychiatry. 2009;66(3):275–285.CrossRefGoogle ScholarPubMed
Baumel, WT, Lu, L, Huang, X, et al. Neurocircuitry of treatment in anxiety disorders. Biomark Neuropsychiatry. 2022;6:100052.10.1016/j.bionps.2022.100052CrossRefGoogle ScholarPubMed
Figure 0

Figure 1. Preferred reporting items for systematic reviews and meta-analyses flowchart of literature search and selection criteria.

Figure 1

Table 1. Summary of the Characteristics of Studies Included in the Meta-Analysis

Figure 2

Table 2. Brain Activations during Negative Emotion Processing in Patients with Anxiety Disorders Compared to Controls

Figure 3

Figure 2. Activations during negative emotion processing in patients with anxiety disorders compared to controls. The red clusters indicate increased activation (hyperactivation); the blue clusters indicate decreased activation (hypoactivation). Abbreviations: ACC, anterior cingulate cortex; B, bilateral; dlPFC, dorsolateral prefrontal cortex; dmPFC, dorsomedial prefrontal cortex; L, left; MTG, middle temporal gyrus; PCC, posterior cingulate cortex; R, right; SPG, superior parietal gyrus; STG, superior temporal gyrus; vlPFC, ventrolateral prefrontal cortex.

Figure 4

Figure 3. The results of meta-regression analysis. Abbreviations: MTG, middle temporal gyrus; PCC, posterior cingulate cortex; STG, superior temporal gyrus.

Supplementary material: File

Wang et al. supplementary material

Wang et al. supplementary material
Download Wang et al. supplementary material(File)
File 736.7 KB