Is the representation of fear distributed across the whole brain?

Is the representation of fear distributed across the whole brain? | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Is the representation of fear distributed across the whole brain? Vincent Taschereau-Dumouchel, Marjorie Côté, Darius Valevicius, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9456965/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract The neural basis of the subjective experience of fear remains incompletely understood. Although fear has traditionally been associated with a circumscribed set of brain regions, recent findings have challenged this view by suggesting that fear-related representations may be widely distributed across the entire brain. In the present study, we investigate the validity of this claim by testing whether a common neural network can be reliably identified across three independent functional MRI datasets (total N = 251) probing distinct fear domains: (1) fear of personally relevant animals, (2) fear of diverse threatening images and (3) fear of conditioned threat cues. Using a combination of mass-univariate analyses and multivariate pattern analysis, we identified a core network consistently engaged across all three datasets, encompassing the anterior cingulate gyrus, the insular cortex, and lateral occipitotemporal regions. Notably, while fear could be predicted from a distributed pattern within each dataset, a shared network could be identified across datasets and analytical approaches. These findings suggest that fear representations are neither strictly localized nor fully diffuse, but instead rely on a reproducible set of regions embedded within broader brain-wide patterns. We discuss how methodological decisions influence conclusions about the spatial organization of fear in the brain, and consider the implications of these results for translational and clinical efforts aimed at modulating pathological fear. Health sciences/Diseases/Psychiatric disorders Biological sciences/Neuroscience Subjective fear fMRI Mass-univariate analyses MVPA canonical system Figures Figure 1 Figure 2 Figure 3 Introduction Fear is a fundamental emotional state that supports survival by enabling organisms to detect, evaluate, and respond to potential threats. 1 Despite its central role in adaptive behavior, and its clear implication in anxiety disorders and phobias, there remains substantial debate regarding how subjective fear is represented in the human brain. 2 While decades of neuroimaging research have identified a set of regions consistently engaged during fear-related tasks, the extent to which these regions constitute a coherent and generalizable neural representation of subjective fear remains unresolved. 3 – 8 Meta-analytic and mass-univariate studies have converged on a so-called canonical fear network, typically encompassing the amygdala, anterior cingulate cortex, insula, hippocampus, and prefrontal regions. 4 , 9 – 15 This network overlaps considerably with broader systems involved in negative affect and salience processing, suggesting that fear may not rely on a strictly specialized neural substrate. Yet, these approaches primarily emphasize localized activations and may underestimate the contribution of distributed neural signals that jointly encode subjective experience. 16 – 18 Recent advances in multivariate pattern analysis (MVPA) have challenged this canonical view by modelling fear as a distributed pattern of activity spanning multiple brain regions simultaneously. 5 , 19 Using MVPA, a previous study reported that subjective fear is better predicted by activity in prefrontal, occipital, and ventral temporal cortices, whereas physiological threat responses are more closely linked to subcortical and interoceptive regions such as the amygdala and insula. 5 These findings have been interpreted as evidence for a dissociation between neural systems supporting threat detection and those supporting the conscious experience of fear. However, this interpretation has been challenged by Zhou and colleagues, who reported that subjective fear could be predicted equally well using randomly sampled voxels distributed across the entire brain as by any specific functional network. 20 On this basis, the authors argued that subjective fear may lack a privileged neural locus and instead emerge from highly distributed brain-wide representations. This conclusion is rather surprising as their own mass-univariate analyses revealed a network that substantially overlapped with the canonical fear system (Sup. Figure 3a in Zhou et al., 2021). This apparent discrepancy raises a critical question: do MVPA results indicate that the notion of a canonical fear network is fundamentally wrong, or do they rather reflect differences in how neural information is captured and evaluated by those two different statistical approaches? One possibility is that MVPA can leverage subtle activation patterns distributed across many brain regions, including those outside traditional emotion networks. 18 While these patterns may improve prediction, they might reflect correlated processes (e.g., attention, arousal) rather than fear representation per se. 21 Another possibility is that information captured outside the canonical network reflects task-specific or dataset-specific features that correlate with fear ratings but do not generalize across paradigms. Another consideration pertains to the studied networks. Zhou et al.’s analyses tested predefined resting-state and “consciousness” networks but not the canonical fear network itself, leaving open the question of whether fear-related information is meaningfully concentrated within regions previously identified by mass-univariate meta-analyses. 22 , 23 Clarifying this issue is particularly important for translational research. If subjective fear is encoded in a stable and generalizable neural network, this network may represent a viable target for clinical interventions. Conversely, if fear-related information is highly distributed and largely context-dependent, neural markers derived from MVPA may show limited generalizability despite strong within-sample performance. In the present study, we directly address this issue by combining mass-univariate and multivariate approaches across three independent fMRI datasets capturing distinct forms of subjective fear: fear of personally relevant animals, fear of diverse threatening images, and fear of conditioned threat cues. By examining whole-brain decoding performance, voxel sampling strategies, and region-wise predictive capacity, we aim to determine whether subjective fear is primarily encoded within a canonical neural network, and whether distributed information outside this network reflects core, generalizable representations or paradigm-specific information. Materials and Methods Participants and experimental design We leverage three functional magnetic resonance imaging (fMRI) datasets that collected participants' subjective fear ratings: (1) the Animal Fear Schema Dataset (AFSD; n = 31), in which participants with self-reported elevated fear of specific animals were presented with a series of images of animals; (2) the Visually Induced Fear Dataset (VIFD; n = 67), which includes the presentation of diverse threatening stimuli (scenes, animals, objects) presented to healthy volunteers screened for psychiatric disorders; and (3) the Conditioned Fear Dataset (CoFD; n = 153), which includes healthy volunteers screened for psychiatric/medical disorders undergoing a classical fear conditioning paradigm, where participants were invited to provide their subjective report of shock likelihood. 5 , 20 , 24 For CoFD, we analyzed extinction, renewal and recall phases rather than initial acquisition to avoid confounding subjective fear with pain responses to shock delivery. During extinction (Day 1) and renewal (Day 2), no shocks were administered, allowing us to isolate learned fear representations independent of nociceptive processing. This approach ensures that shock expectancy ratings reflect anticipated threat rather than responses to actual aversive stimuli. See Fig. 1 for more details on experimental protocols. Data preprocessing Preprocessing for AFSD and VIFD is described in the original publications. 5 , 20 For CoFD, first-level GLMs were computed using SPM12 with motion regressors and high-pass filtering (128s cutoff). We excluded 22 participants due to missing data, technical errors, or generalized fear responses (high fear for CS-). For each dataset, fear ratings were combined with neuroimaging data and averaged within-subject across fear levels, yielding beta files for up to six subjective fear levels per participant. See supplementary data for more information. Datasets were spatially aligned and mean-centered within participants. Analyses Mass-univariate analyses Mass univariate analyses are the typical approach for analyzing localized neural activations in relation to an experimental stimulus. 17 In contrast to MVPA, mass univariate analyses assess the relationship between each voxel and the experimental variable independently, typically followed by multiple-comparison correction, such as false discovery rate (FDR) control. 25 , 26 Usually, this takes the form of a generalized linear model (GLM), which estimates betas (slopes) and t-statistics between the time series of each voxel in the brain and a representation of the experimental variable which has been convolved with a canonical haemodynamic response function (HRF), while accounting for other sources of noise such as scanner drift, motion, and physiological noise. 27 However, for two of the three studies, data were already reduced to a single beta map per fear level and per participant. As we did not have access to the time-series data itself, we opted for a simplified mass-univariate analysis, pooling subject data within studies and fitting multilevel regression models (MLMs) for each voxel, with subject ID as a grouping factor and a random intercept. The independent fixed-effects variable was the voxel's beta value, and the dependent variable was fear level. This produced one spatial map of beta coefficients and p-values for each of the three datasets, which were then combined using Fisher's z-transform method. The p-values were then adjusted for multiple comparisons using an FDR correction. The results are presented in Table 1 (with adjusted p-values averaged by Brainnetome ROIs) and Fig. 2 A. Additionally, we computed Pearson correlations to visualize cross-dataset agreement (Fig. 2 B-C). Table 1 Brain regions showing significant associations with subjective fear ratings in combined mass-univariate analysis (AFSD, VIFD, CoFD; N = 251). Anatomical labels from Brainnetome Atlas; β-values from combined MLM analysis; p-values combined via Fisher's z-transform. See text for details. Lobe Gyrus Region Hemisphere b-value FDR-corrected p-value Frontal lobe SFG, Superior Frontal Gyrus A8m, medial area 8 Left 0.08 0.04 A6dl, dorsolateral area 6 Left 0.05 0.047 MFG, Middle Frontal Gyrus A10l, lateral area10 Right -0.06 0.025 IFG, Inferior Frontal Gyrus A44d,dorsal area 44 Left 0.06 0.043 A44d,dorsal area 44 Right 0.08 0.001 A45c, caudal area 45 Right 0.06 0.004 A44op, opercular area 44 Left 0.08 0.013 A44op, opercular area 44 Right 0.1 0.001 PrG, Precentral Gyrus A6cvl, caudal ventrolateral area 6 Left 0.08 0.044 Temporal Lobe MFG, Middle Temporal Gyrus A37dl, dorsolateral area37 Right 0.08 0.024 ITG, Inferior Temporal Gyrus A37elv, extreme lateroventral area37 Left 0.1 0.005 A37vl, ventrolateral area 37 Left 0.11 0.025 A37vl, ventrolateral area 37 Right 0.12 0.004 FuG, Fusiform Gyrus A37mv, medioventral area37 Left 0.12 < 0.001 A37mv, medioventral area37 Right 0.1 0.005 A37lv, lateroventral area37 Left 0.12 0.015 A37lv, lateroventral area37 Right 0.12 0.004 PhG, Parahippocampal gyrus A35/36r, rostral area 35/36 Left 0.07 0.01 A35/36r, rostral area 35/36 Right 0.04 0.043 TI, area TI(temporal agranular insular cortex) Left 0.06 0.012 pSTS, Posterior Superior Temporal cpSTS, caudoposterior superior temporal sulcus Left 0.08 0.007 cpSTS, caudoposterior superior temporal sulcus Right 0.08 0.023 Parietal Lobe SPL, Superior Parietal Lobule A7r, rostral area 7 Left 0.06 0.002 A7c, caudal area 7 Left 0.06 0.006 A5l, lateral area 5 Left 0.08 0.016 A5l, lateral area 5 Right 0.06 0.046 A7ip, intraparietal area 7(hIP3) Left 0.07 0.009 A7ip, intraparietal area 7(hIP3) Right 0.07 0.027 IPL, Inferior Parietal Lobule A40rd, rostrodorsal area 40(PFt) Left 0.08 0.031 Insular Lobe INS, Insular Gyrus vIa, ventral agranular insular Left 0.1 0.001 vIa, ventral agranular insular Right 0.09 0.009 dIa, dorsal agranular insular Left 0.14 < 0.001 dIa, dorsal agranular insular Right 0.12 0.001 dId, dorsal dysgranular insular Left 0.1 0.016 dId, dorsal dysgranular insular Right 0.1 0.044 Limbic Lobe CG, Cingulate Gyrus A24rv, rostroventral area 24 Left 0.08 0.025 A32p, pregenual area 32 Left 0.1 0.008 A32p, pregenual area 32 Right 0.1 0.028 A24cd, caudodorsal area 24 Left 0.13 0.001 A24cd, caudodorsal area 24 Right 0.12 0.001 A23c, caudal area 23 Left 0.06 0.033 Occipital Lobe LOcC, Lateral Occipital Cortex mOccG, middle occipital gyrus Left 0.08 0.028 V5/MT+, area V5/MT+ Left 0.12 < 0.001 V5/MT+, area V5/MT+ Right 0.1 < 0.001 iOccG, inferior occipital gyrus Left 0.07 0.034 iOccG, inferior occipital gyrus Right 0.07 0.013 lsOccG, lateral superior occipital gyrus Left 0.08 0.002 Subcortical Nuclei Amyg, Amygdala mAmyg, medial amygdala Left 0.11 < 0.001 mAmyg, medial amygdala Right 0.09 0.014 lAmyg, lateral amygdala Left 0.08 0.009 lAmyg, lateral amygdala Right 0.08 0.015 BG, Basal Ganglia GP, globus pallidus Right 0.06 0.031 vmPu, ventromedial putamen Left 0.06 0.015 vmPu, ventromedial putamen Right 0.06 0.032 Tha, Thalamus mPFtha, medial pre-frontal thalamus Left 0.12 0.001 mPFtha, medial pre-frontal thalamus Right 0.09 0.002 mPMtha, pre-motor thalamus Right 0.1 0.008 Stha, sensory thalamus Right 0.06 0.04 PPtha, posterior parietal thalamus Left 0.07 0.038 lPFtha, lateral pre-frontal thalamus Left 0.1 0.015 lPFtha, lateral pre-frontal thalamus Right 0.09 0.008 Whole-brain decoding performance across datasets To predict subjective fear ratings from whole-brain activation patterns, we applied Support Vector Regression (SVR; CANlab toolbox, https://github.com/canlab/CanlabCore ) with leave-one-subject-out cross-validation to each dataset and to the combined datasets. Model performance was evaluated using Pearson correlations between observed and predicted fear ratings, and area under the ROC curve (AUC) for discriminating low-fear (0–2) from high-fear (3–5) trials derived from continuous SVR outputs. Statistical significance was assessed via 1,000 permutations. To examine cross-dataset transferability, models trained on each dataset were applied to the remaining two datasets, with performance evaluated using both metrics and FDR-corrected p-values across all cross-prediction tests. Negative Affect Mask Decoder To test Zhou et al.'s proposition that fear lacks a unique neural locus, we implemented a voxel-sampling approach examining how prediction accuracy scales with brain coverage. 20 We trained SVR models on: 1) randomly selected voxel subsets (100 to 348,904 voxels; 15 iterations per subset), and 2) voxels constrained to Lindquist et al.'s meta-analytic "negative affect" mask available on Neurosynth (negative affect_uniformity-test_z_FDR.01.nii). 28 This allowed direct comparison of decoder performance within the canonical network versus random, spatially distributed voxels. Performance was measured using Pearson correlations between observed and predicted fear ratings, with FDR correction across six model comparisons (3 datasets × 2 conditions: whole-brain vs. mask). ROI Analysis Next, we analyzed the predictive capacity of the decoders from each dataset to predict subjective reports across 214 regions of the Brainnetome Atlas, a brain parcellation atlas based on brain connectivity, including 210 cortical regions and four subcortical regions (bilateral amygdala and hippocampi). 29 For each Brainnetome ROI, we extracted the voxel data by applying the corresponding mask and predicted the subjective ratings using a Support Vector Regression (SVR) model in a leave-one-subject-out cross-validation procedure. Model performance was assessed by computing Pearson correlations between real and predicted fear ratings. To statistically compare correlation coefficients across datasets, we applied Fisher's z-transformation, which converts correlation coefficients into z-scores, enabling statistical comparison between datasets. 30 We assessed the significance of correlation coefficients within each dataset by correcting for multiple comparisons across ROIs using the False Discovery Rate (FDR) procedure (Benjamini-Hochberg, p < 0.05, assuming dependent tests). Results Mass-univariate analyses Combining the three datasets (AFSD, VIFD, and CoFD) revealed significant positive activations (red and yellow) in the prefrontal cortex (superior and inferior frontal gyrus), the precentral gyrus, the postcentral gyrus, the anterior cingulate and insular gyrus, and subcortical structures including the amygdala, basal ganglia and thalamus (Fig. 2A and Table 1). Additional positive activations were observed in the temporal (inferior, middle, fusiform, parahippocampal and posterior temporal gyrus), parietal (superior and inferior lobules) and lateral occipitotemporal cortex. Negative activations (blue) were primarily located in the middle frontal gyrus (Table 1). Overall, these results indicate that fear is represented in a network including the prefrontal, temporal, occipital, cingulate, and subcortical regions, alongside decreased activity in medial prefrontal areas. The overlap map highlights the degree of consistency across datasets (Fig. 2B). Most voxels showing significant activations in all three datasets were located in the anterior cingulate gyrus, insula, diencephalon, mesencephalon, genu of the corpus callosum, and lateral occipitotemporal cortex (yellow clusters). Areas unique to a single dataset (blue) were primarily confined to occipital and inferior temporal cortices, suggesting potential task- or stimulus-related variability. Peak coordinates and statistical values for all significant clusters are reported in Table 1. Whole-brain decoding performance across datasets We first examined whether subjective fear ratings could be predicted from whole-brain activation patterns across all three datasets. Support Vector Regression models with leave-one-subject-out cross-validation revealed significant above-chance prediction accuracy for all datasets (Fig. 3 and Table 2). The AFSD showed the strongest decoding performance (r = 0.749, p < 0.001; AUC = 0.880, p = 0.002), followed by VIFD (r = 0.574, p < 0.001; AUC = 0.795, p = 0.002) and CoFD (r = 0.447, p < 0.001; AUC = 0.750, p = 0.002). Whole-brain activation maps revealed that subjective fear was associated with distributed neural responses across the cortex in all three datasets (see Fig. 3C). Rather than being restricted to specific emotion-related regions, activation patterns were widespread, involving occipital, temporal, parietal, and frontal areas. Table 2: Correlation of subjective fear judgments with predicted fear from brain activity according to each dataset Dataset Pearson Correlation (r) corrected p-value ( p ) Area Under the Curve (AUC) corrected p-value ( p ) AFSD 0.749* < 0.001 0.880* 0.002 VIFD 0.574* < 0.001 0.795* 0.002 CoFD 0.447* < 0.001 0.750* 0.002 AFSD( VIFD ) 1 0.627* < 0.001 0.803* 0.002 AFSD( CoFD ) 1 0.285* < 0.001 0.601* 0.009 VIFD( AFSD ) 1 0.182* < 0.001 0.608* 0.003 VIFD( CoFD ) 1 0.191* < 0.001 0.581* 0.002 CoFD( AFSD ) 1 0.098* 0.025 0.565* 0.003 CoFD( VIFD ) 1 0.190* < 0.001 0.633* 0.002 1 Cross-prediction to see if brain patterns learned in one model dataset can be generalized to another dataset where “Data( Model )”. * = p < 0.05 after FDR correction. Cross-dataset generalization analyses To assess whether fear representations generalized across datasets, we applied decoders trained on one dataset to predict fear ratings in the others (Table 2). Generally, models showed modest generalizability to other datasets. Models trained on VIFD showed strong generalization to AFSD (r = 0.627, p < 0.001; AUC = 0.803, p = 0.002) and weak generalization to CoFD (r = 0.190, p < 0.001; AUC = 0.633, p = 0.002). Conversely, models trained on AFSD generalized weakly to VIFD (r = 0.182, p < 0.001; AUC = 0.608, p = 0.003) and to CoFD (r = 0.098, p = 0.025; AUC = 0.565, p = 0.003). Models trained on CoFD similarly showed weak generalization to both AFSD (r = 0.285, p < 0.001; AUC = 0.601, p = 0.003) and VIFD (r = 0.191, p < 0.001; AUC = 0.591, p = 0.002). Restricting predictions to the canonical fear mask Figure 3B shows Pearson correlations between observed and predicted fear ratings across all three datasets as the number of randomly selected voxels increased from 100 to 348,904 voxels. When we restricted analyses to voxels within the Lindquist meta-analytic mask, we observed distinct patterns across datasets. In VIFD and COFD, constraining predictions to the canonical emotion network decreased performance relative to random voxel selection, suggesting that the decoders likely rely on information outside these regions to make predictions. In contrast, for AFSD, prediction accuracy within the mask closely matched that of random selection, suggesting that little unique information is represented outside the canonical network in this dataset. Prediction accuracy using all brain voxels was significant for all three datasets (see Table 2 or Fig. 3A for detailed statistics). When restricting analyses to the canonical fear mask, performance remained significant and showed similar patterns (AFSD: r = 0.681, p < 0.001; VIFD: r = 0.358, p < 0.001; CoFD: r = 0.274, p < 0.001). At approximately 10,000 randomly selected voxels, performance reached a plateau, and mean correlations were 0.746 ± 0.008 (SD) for AFSD, 0.556 ± 0.013 for VIFD, and 0.426 ± 0.021 for CoFD. When restricting analyses to the canonical fear mask, performance also plateaued at approximately 10,000 randomly selected voxels, and mean correlations were 0.681 ± 0.009 (SD) for AFSD, 0.355 ± 0.010 for VIFD, and 0.264 ± 0.014 for CoFD. The decoder trained on the three combined datasets failed to reach statistical significance (p > 0.05), indicating no reliable predictive performance. Common regions for the prediction of fear across datasets Region-by-region analyses across 214 brain regions from the Brainnetome Atlas identified 23 common regions that showed significant brain-subjective fear correlations across all three datasets (Pearson r = 0.182–0.306; FDR-corrected p < 0.05 within each dataset; Figure 3D, Table 3). These regions include anterior cingulate gyrus, insular cortex (agranular insular), visual processing areas (lateral and medioventral occipital cortex), memory-related structures (parahippocampal gyrus), temporal cortex (fusiform and inferior temporal gyri), orbitofrontal cortex (orbital gyrus), parietal regions (precuneus), sensorimotor areas (paracentral lobule), prefrontal cortex (superior and inferior frontal gyri), and subcortical nuclei (amygdala). The strongest mean correlations were observed in left area V5/MT+ of the lateral occipital cortex (r = 0.306), right medioventral area 37 of the fusiform gyrus (r = 0.300), and left pregenual area 32 of the cingulate gyrus (r = 0.286). Table 3. Regions with significant brain-behaviour correlations across all datasets (N = 251; FDR-corrected q < 0.05). SVR predictions with leave-one- subject-out cross-validation. Mean r and p-values across datasets shown. Lobe Gyrus Region Hemisphere r-value (mean) FDR-corrected p-value (mean) Frontal lobe SFG, Superior Frontal Gyrus A8m, medial area 8 Right 0.204 0.001 A9l, lateral area 9 Right 0.212 <0.001 IFG, Inferior Frontal Gyrus A45r, rostral area 45 Left 0.200 0.001 A44v, ventral area 44 Left 0.182 0.002 OrG, Orbital Gyrus A11l, lateral area 11 Right 0.269 <0.001 A12/47l, lateral area 12/47 Right 0.270 <0.001 PCL, Paracentral Lobule A1/2/3ll, area 1/2/3 (lower limb) Right 0.213 <0.001 Temporal Lobe ITG, Inferior Temporal Gyrus A20cv, caudoventral area 20 Left 0.252 <0.001 FuG, Fusiform Gyrus A37mv, medioventral area37 Right 0.300 <0.001 A37lv, lateroventral area37 Right 0.264 <0.001 PhG, Parahippocampal gyrus TI, area TI Left 0.285 <0.001 TH, area TH Right 0.278 <0.001 Parietal Lobe Pcu, Precuneus A5m, medial area 5 Left 0.264 <0.001 A5m, medial area 5 Right 0.226 <0.001 A31m, medial area 31 Right 0.217 0.002 Insular Lobe INS, Insular Gyrus vIa, ventral agranular insular Left 0.225 0.001 dIa, dorsal agranular insular Right 0.243 <0.001 Limbic Lobe CG, Cingulate Gyrus A32p, pregenual area 32 Left 0.286 <0.001 Occipital Lobe MVOcC, Medioventral Occipital Cortex rLingG, rostral lingual gyrus Right 0.282 0.001 vmPOS, ventromedial parietooccipital sulcus Right 0.270 0.002 LOcC, Lateral Occipital Cortex V5/MT+, area V5/MT+ Left 0.306 <0.001 iOccG, inferior occipital gyrus Left 0.247 <0.001 Subcortical Nuclei Amyg, Amygdala lAmyg, lateral amygdala Left 0.212 0.001 Mass-univariate analyses and MVPA combined When comparing our two primary analyses (mass-univariate and MVPA), we found eight regions that were consistently identified across all three datasets: the fusiform gyrus (right medioventral and lateroventral area 37; b = 0.1 and b = 0.12; r mean = 0.300 and r mean = 0.264), the insula (left ventral agranular and right dorsal dysgranular; b = 0.1 and b = 0.12; r mean = 0.225 and r mean = 0.243), the cingulate gyrus (right pregenual area 32; b = 0.1; r mean = 0.286), the occipital gyrus (left V5/MT+ and left inferior occipital gyrus; b = 0.12 and b = 0.07; r mean = 0.306 and r mean = 0.247), and the left lateral amygdala (b = 0.08; r mean = 0.212) (see Table 1 and 3). Discussion This study addressed an important question in affective neuroscience: Is subjective fear represented in a localized canonical network or distributed across the entire brain? 20 By combining mass-univariate and multivariate pattern analyses across three independent fMRI datasets (N = 251 total) capturing distinct fear contexts (animal-specific fear, visually-induced situational fear, and conditioned fear cues), we provide evidence for both perspectives. Our key findings are threefold. First, mass-univariate analyses and MVPA identified a consistent network spanning the anterior cingulate cortex, insula, fusiform gyrus, lateral occipital cortex (V5/MT+), and amygdala. Second, whole-brain MVPA demonstrated that distributed patterns significantly predict subjective fear, consistent with Zhou et al.'s findings. Third, critically, restricting analyses to a meta-analytic negative affect mask did not improve prediction over random voxel selection, yet the eight convergent regions exhibited both robust focal activation and informative multivariate patterns. These findings reconcile apparently contradictory perspectives by demonstrating that a canonical network exists but does not contain all fear-relevant information, and that distributed information outside this network appears to be primarily paradigm-specific rather than generalizable. The canonical network of fear: functional specialization and integration The eight consistently identified regions from mass-univariate and MVPA approaches form a functionally coherent network integrating threat detection, interoceptive awareness, visual processing, and conscious emotional experience. Amygdala and threat detection. The presence of the left lateral amygdala is consistent with its established role in threat detection and associative learning, with recent evidence suggesting the left amygdala specifically mediates cognitively-mediated fear acquisition and extinction. 11 , 31 Its recruitment across fear conditions underscores its function as a rapid and reliable threat detector, operating early in the fear processing hierarchy to initiate downstream cortical and subcortical responses. Insula and interoceptive integration . Bilateral insula activation likely reflects the integration of bodily signals (heart rate, muscle tension, visceral sensations) 32 The insula's role as an interoceptive hub makes it well-positioned to transform physiological arousal into subjective feeling states. Its consistent engagement across paradigms suggests that subjective fear requires awareness of bodily changes, regardless of whether the trigger is a phobic animal, threatening scene, or conditioned stimulus. Anterior cingulate cortex and emotional expression . The pregenual ACC's involvement aligns with its documented role in cognitive control, conflict monitoring, and emotion-related appraisal, including fear evaluation. 33 This region may adjudicate between competing interpretations of ambiguous stimuli and modulate the intensity of fear expression. Its activation could reflect the evaluative component of fear: "How threatening is this stimulus to me?" Fusiform gyrus as a visual-emotional interface. The fusiform gyrus is broadly involved in high-level visual processing, including fine-grained object discrimination, perceptual expertise, and category-sensitive representations. 34 Through its connectivity with the anterior temporal lobe, which supports semantic memory, and with medial temporal structures such as the hippocampus, the fusiform gyrus is well positioned to integrate visual representations with mnemonic and emotional context. 35 – 37 Fearful stimuli, whether animals, threatening scenes, or conditioned cues, may amplify perceptual processing within ventral occipito-temporal cortex when they are motivationally relevant, a process that likely draws on these multimodal connections. 38 Importantly, a multivoxel decoding study suggests that ventral occipito-temporal regions contribute significantly to predicting subjective fear ratings, and in some cases, fusiform activity patterns predict self-reported fear more strongly than peripheral autonomic measures, suggesting that these representations are more closely linked to conscious experience than to purely defensive responses. 5 In our data, the right medioventral and lateroventral fusiform (area 37) exhibited among the strongest and most consistent cross-dataset associations in both mass-univariate and ROI decoding analyses, reinforcing this interpretation. This interpretation is consistent with evidence suggesting that motivationally relevant stimuli, including fear-inducing ones, may recruit high-level visual mechanisms. 39 – 42 When stimuli are fear-inducing, their visual features may thus be encoded with greater resolution or gain, increasing their subjective salience. This is consistent with evidence showing that emotional salience and attention jointly modulate fusiform responses, and that fusiform–amygdala connectivity increases in anxiety states. 43 The fusiform gyrus may therefore act as a visual–emotional amplification hub, where perceptual representations become tightly coupled with affective evaluation. 44 Lateral occipital cortex and threat feature extraction. The lateral occipital cortex, encompassing area V5/MT + and the inferior occipital gyrus, is classically associated with motion processing, visual feature extraction, and the perception of biological movement. 45 , 46 Notably, the engagement of V5/MT+ across all three paradigms, including conditioned light cues devoid of any inherent motion, suggests that its activation is unlikely to reflect low-level stimulus features alone. Instead, it may reflect top-down modulation driven by the learned or intrinsic motivational relevance of the stimulus, consistent with evidence that the amygdala can amplify activity in motion-sensitive visual regions in the absence of actual movement. 47 Beyond motion per se, the inferior occipital gyrus likely contributes lower-level feature extraction that feeds into the higher-level representations computed in fusiform and parahippocampal regions. 48 Together, these lateral occipital regions may constitute an early visual gateway for threat prioritization, rapidly extracting stimulus features that are subsequently elaborated by downstream temporal and limbic structures. 49 – 51 Consistent with this view, prior work demonstrated that the ventral visual stream encodes affectively neutral visual features statistically associated with fear ratings, independently of whether subjective fear is actually reported, and that it is the multivariate information transmission between ventral visual areas and prefrontal regions that distinguishes participants reporting subjective fear from those who do not. 38 Distributed representations: separating signal from noise Our voxel-sampling analyses replicate Zhou et al.'s finding that prediction improves with increasing brain coverage, seemingly supporting the view that subjective fear lacks a privileged neural locus. However, we made three critical observations: First, cross-dataset generalization was weak (r < 0.30 for most comparisons), indicating that whole-brain decoders primarily capture paradigm-specific variance rather than generalizable fear representations. A truly distributed, content-general fear representation should transfer robustly across contexts and our results suggest otherwise. Second, restricting analyses to a meta-analytic negative affect network did not enhance prediction in VIFD and CoFD, confirming that decoders leverage information outside canonical emotion regions. Only AFSD showed comparable performance within versus outside this network, possibly reflecting stronger canonical circuit engagement in consolidated fear schemas relative to situational or conditioned fear. Third, the combined-dataset decoder failed to reach significance (p > 0.05): if fear were truly represented in a universal distributed manner, pooling datasets should enhance power, but instead obscures shared canonical signals, suggesting that much of the distributed signal may reflect idiosyncratic task demands rather than core fear processes. Collectively, these findings suggest a hybrid model: a canonical network reliably represents core fear components across contexts, while distributed patterns may capture context-specific elaborations that improve within-sample prediction but generalize poorly across datasets. Reconciling seemingly contradictory approaches The apparent contradiction between mass-univariate and MVPA findings is best understood as a reflection of each approach's design rather than a fundamental disagreement about the neural basis of fear. Mass-univariate analyses detect locally robust, spatially focal activations that replicate across paradigms, whereas MVPA exploits multivariate spatial patterns that may improve within-sample prediction by leveraging task-specific covariance structures. 52 , 53 These approaches therefore operate on different sources of neural variance, and their convergence, rather than their divergence, is most informative. The eight regions identified by both methods provide the strongest evidence for a canonical network, exhibiting both robust focal activation and informative patterns across all three datasets. Critically, the failure of the combined-dataset decoder (p > 0.05) further supports this interpretation: pooling datasets without accounting for task-specific variance obscures shared canonical signals, underscoring that the generalizable component of fear-related neural activity is concentrated within a reproducible set of regions rather than diffusely distributed across the brain. 54 Methodological considerations and limitations Several limitations should be noted. Gender was anonymized in VIFD, limiting our ability to examine sex differences in fear processing. 55 , 56 Experimental designs differed substantially across datasets in presentation duration, rating timing, trial numbers (720 vs. 80 vs. 16 analyzed trials), and rating scales. These differences may explain varying decoding performance and limited cross-dataset generalization. Furthermore, we relied on a single dataset per paradigm type, limiting our ability to distinguish generalizable from dataset-specific features. Future work should replicate findings across multiple datasets per paradigm type or employ within-subject designs that combine multiple fear paradigms (e.g., specific, general, and conditioned fear). Conclusion Our findings support the hypothesis of a canonical fear network comprising the fusiform gyrus, insula, anterior cingulate cortex, lateral occipital cortex, and amygdala. The convergence of mass-univariate and MVPA approaches demonstrates that methodological choices are critical. Mass-univariate analyses may be tailored to identify robust, generalizable focal effects, while MVPA may capture additional information that may be task-specific. Considering these methodological issues is essential for translational neuroscience seeking to identify neural predictors of treatment response. These findings have important clinical implications for anxiety disorders and phobias. The canonical fear network identified here provides reliable neural targets for therapeutic intervention, where intervention targeting the modulation of these regions might yield consistent therapeutic effects. Future studies combining multiple fear paradigms within participants are recommended to distinguish methodological from experiential variance. Declarations Conflicts of interest Authors declare that there are not any competing financial interests in relation to the work described. Acknowledgements Marjorie Côté was supported by the Natural Sciences and Engineering Research Council of Canada and Dr Vincent Taschereau-Dumouchel was supported in part by the Fond de recherche du Québec - Santé and the Fondation de l’Institut universitaire en santé mentale de Montréal. References Mobbs, D. The ethological deconstruction of fear(s). Curr. Opin. Behav. Sci. 24, 32–37 (2018). Garcia, R. Neurobiology of fear and specific phobias. Learn. Mem. 24, 462–471 (2017). LeDoux, J. & Brown, R. A higher-order theory of emotional consciousness. Proc. Natl. Acad. Sci. 114, E2016–E2025 (2017). LeDoux, J. & Pine, D. S. Using Neuroscience to Help Understand Fear and Anxiety: A Two-System Framework. Am. J. Psychiatry 173, 1083–1093 (2016). Taschereau-Dumouchel, V., Kawato, M. & Lau, H. Multivoxel pattern analysis reveals dissociations between subjective fear and its physiological correlates. Mol. Psychiatry 25, 2342–2354 (2020). Taschereau-Dumouchel, V., Michel, M., Lau, H., Hofmann, S. G. & LeDoux, J. E. Putting the ‘mental’ back in ‘mental disorders’: a perspective from research on fear and anxiety. Mol. Psychiatry 27, 1322–1330 (2022). Wendt, J., Löw, A., Weymar, M., Lotze, M. & Hamm, A. O. Active avoidance and attentive freezing in the face of approaching threat. NeuroImage 158, 196–204 (2017). Mauss, I. B. & Robinson, M. D. Measures of emotion: A review. Cogn. Emot. 23, 209–237 (2009). Gilmartin, M. R., Balderston, N. L. & Helmstetter, F. J. Prefrontal cortical regulation of fear learning. Trends Neurosci. 37, 455–464 (2014). Milad, M. R. et al. Recall of fear extinction in humans activates the ventromedial prefrontal cortex and hippocampus in concert. Biol. Psychiatry 62, 446–454 (2007). Moustafa, A. A. et al. A model of amygdala–hippocampal–prefrontal interaction in fear conditioning and extinction in animals. Brain Cogn. 81, 29–43 (2013). Picó-Pérez, M. et al. Common and distinct neural correlates of fear extinction and cognitive reappraisal: A meta-analysis of fMRI studies. Neurosci. Biobehav. Rev. 104, 102–115 (2019). Ressler, K. J. Translating Across Circuits and Genetics Toward Progress in Fear- and Anxiety-Related Disorders. Am. J. Psychiatry 177, 214–222 (2020). Chikazoe, J., Lee, D. H., Kriegeskorte, N. & Anderson, A. K. Population coding of affect across stimuli, modalities and individuals. Nat. Neurosci. 17, 1114–1122 (2014). Lindquist, K. A., Satpute, A. B., Wager, T. D., Weber, J. & Barrett, L. F. The Brain Basis of Positive and Negative Affect: Evidence from a Meta-Analysis of the Human Neuroimaging Literature. Cereb. Cortex N. Y. N 1991 26, 1910–1922 (2016). Wen, Z. et al. Distributed neural representations of conditioned threat in the human brain. Nat. Commun. 15, 2231 (2024). Mahmoudi, A., Takerkart, S., Regragui, F., Boussaoud, D. & Brovelli, A. Multivoxel Pattern Analysis for fMRI Data: A Review. Comput. Math. Methods Med. 2012, 961257 (2012). Weaverdyck, M. E., Lieberman, M. D. & Parkinson, C. Tools of the Trade Multivoxel pattern analysis in fMRI: a practical introduction for social and affective neuroscientists. Soc. Cogn. Affect. Neurosci. 15, 487–509 (2020). Kragel, P. A. & LaBar, K. S. Advancing emotion theory with multivariate pattern classification. Emot. Rev. J. Int. Soc. Res. Emot. 6, 160–174 (2014). Zhou, F. et al. A distributed fMRI-based signature for the subjective experience of fear. Nat. Commun. 12, 6643 (2021). Stelzer, J., Chen, Y. & Turner, R. Statistical inference and multiple testing correction in classification-based multi-voxel pattern analysis (MVPA): Random permutations and cluster size control. NeuroImage 65, 69–82 (2013). Buckner, R. L., Krienen, F. M., Castellanos, A., Diaz, J. C. & Yeo., B. T. The organization of the human cerebellum estimated by intrinsic functional connectivity. J. Neurophysiol. 106, 2322–2345 (2011). Yeo, B. T. T. et al. The organization of the human cerebral cortex estimated by intrinsic functional connectivity. J. Neurophysiol. 106, 1125–1165 (2011). Davignon, L.-M. et al. Associations between past and current use of oral contraceptives and fear regulation. Biol. Psychiatry Cogn. Neurosci. Neuroimaging https://doi.org/10.1016/j.bpsc.2025.09.018 (2025) doi:10.1016/j.bpsc.2025.09.018 (2025). Monti, M. M. Statistical Analysis of fMRI Time-Series: A Critical Review of the GLM Approach. Front. Hum. Neurosci. 5, (2011). Genovese, C. R., Lazar, N. A. & Nichols, T. Thresholding of statistical maps in functional neuroimaging using the false discovery rate. NeuroImage 15, 870–878 (2002). Moeller, J. R. & Habeck, C. G. Reciprocal Benefits of Mass-Univariate and Multivariate Modeling in Brain Mapping: Applications to Event-Related Functional MRI, H215O-, and FDG-PET. Int. J. Biomed. Imaging 2006, 079862 (2006). Lindquist, K., Satpute, A., Wager, T., Weber, J. & Barrett, L. Neurosynth: 25631056. https://neurosynth.org/studies/25631056/ (2015). Fan, L. et al. The Human Brainnetome Atlas: A New Brain Atlas Based on Connectional Architecture. Cereb. Cortex N. Y. N 1991 26, 3508–3526 (2016). Bobbitt, Z. Fisher Z-Transformation: Definition & Example. Statology https://www.statology.org/fisher-z-transformation/ (2022). Xie, T. et al. The case for hemispheric lateralization of the human amygdala in fear processing. Mol. Psychiatry 30, 2252–2259 (2025). Gogolla, N. The insular cortex. Curr. Biol. 27, R580–R586 (2017). Stevens, F., Hurley, R. & Taber, K. Anterior Cingulate Cortex: Unique Role in Cognition and Emotion. J. Neuropsychiatry Clin. Neurosci. 23, (2011). Weiner, K. S. & Zilles, K. The anatomical and functional specialization of the fusiform gyrus. Neuropsychologia 83, 48–62 (2016). Mion, M. et al. What the left and right anterior fusiform gyri tell us about semantic memory. Brain J. Neurol. 133, 3256–3268 (2010). Fleury, M. N. et al. Long-term memory plasticity in a decade-long connectivity study post anterior temporal lobe resection. Nat. Commun. 16, 692 (2025). Schwab, S. et al. Functional Connectivity Alterations of the Temporal Lobe and Hippocampus in Semantic Dementia and Alzheimer’s Disease. J. Alzheimers Dis. 76, 1461–1475 (2020). Taschereau-Dumouchel, V. et al. Interaction between the prefrontal and visual cortices supports subjective fear. Philos. Trans. R. Soc. B Biol. Sci. 379, 20230245 (2024). Bilalić, M., Langner, R., Ulrich, R. & Grodd, W. Many faces of expertise: fusiform face area in chess experts and novices. J. Neurosci. Off. J. Soc. Neurosci. 31, 10206–10214 (2011). Tarr, M. J. & Gauthier, I. FFA: a flexible fusiform area for subordinate-level visual processing automatized by expertise. Nat. Neurosci. 3, 764–769 (2000). McGugin, R. W., Gatenby, J. C., Gore, J. C. & Gauthier, I. High-resolution imaging of expertise reveals reliable object selectivity in the fusiform face area related to perceptual performance. Proc. Natl. Acad. Sci. 109, 17063–17068 (2012). McGugin, R. W., Van Gulick, A. E. & Gauthier, I. Cortical Thickness in Fusiform Face Area Predicts Face and Object Recognition Performance. J. Cogn. Neurosci. 28, 282–294 (2016). Jung, Y.-H. et al. Altered Amygdala Resting-State Functional Connectivity and Hemispheric Asymmetry in Patients With Social Anxiety Disorder. Front. Psychiatry 9, 164 (2018). Canchola, E., Matsumura, P. D. & Rivera, R. M. Fusiform Gyrus: The Sorcerer of the Brain and Architect of Subjective Reality. Med. - Clin. - Res. 1, 49–51 (2025). Kryklywy, J. H., Forys, B. J., Vieira, J. B., Quinlan, D. J. & Mitchell, D. G. V. Dissociating representations of affect and motion in visual cortices. Cogn. Affect. Behav. Neurosci. 23, 1322–1345 (2023). Sack, A. T., Kohler, A., Linden, D. E. J., Goebel, R. & Muckli, L. The temporal characteristics of motion processing in hMT/V5+: Combining fMRI and neuronavigated TMS. NeuroImage 29, 1326–1335 (2006). Furl, N., Henson, R. N., Friston, K. J. & Calder, A. J. Top-Down Control of Visual Responses to Fear by the Amygdala. J. Neurosci. 33, 17435–17443 (2013). Palejwala, A. H. et al. Anatomy and white matter connections of the fusiform gyrus. Sci. Rep. 10, 13489 (2020). Lithari, C., Moratti, S. & Weisz, N. Limbic areas are functionally decoupled and visual cortex takes a more central role during fear conditioning in humans. Sci. Rep. 6, 29220 (2016). Li, W. & Keil, A. Sensing fear: fast and precise threat evaluation in human sensory cortex. Trends Cogn. Sci. 27, 341–352 (2023). Bayle, D. J., Henaff, M.-A. & Krolak-Salmon, P. Unconsciously Perceived Fear in Peripheral Vision Alerts the Limbic System: A MEG Study. PLOS ONE 4, e8207 (2009). Davis, T. et al. What do differences between multi-voxel and univariate analysis mean? How subject-, voxel-, and trial-level variance impact fMRI analysis. NeuroImage 97, 271–283 (2014). Hebart, M. N. & Baker, C. I. Deconstructing multivariate decoding for the study of brain function. NeuroImage 180, 4–18 (2018). Wu, J. et al. Cross-cohort replicability and generalizability of connectivity-based psychometric prediction patterns. NeuroImage 262, 119569 (2022). Yeretzian, S. T., Sahakyan, Y., Kozloff, N. & Abrahamyan, L. Sex differences in the prevalence and factors associated with anxiety disorders in Canada: A population-based study. J. Psychiatr. Res. 164, 125–132 (2023). Bauer, E. P. Sex differences in fear responses: Neural circuits. Neuropharmacology 222, 109298 (2023). Additional Declarations The authors have declared there is NO conflict of interest to disclose Supplementary Files S1RDM.pdf Figure S1 S2.pdf Figure S2 S3.pdf Figure S3 S4.pdf Figure S4 S5.pdf Figure S5 S6.pdf Figure S6 SupplementaryData.docx Supplementary Analysis Cite Share Download PDF Status: Under Review Version 1 posted Reviewer # 2 agreed at journal 30 Apr, 2026 Reviewer # 1 agreed at journal 30 Apr, 2026 Reviewers invited by journal 29 Apr, 2026 Editor assigned by journal 21 Apr, 2026 Submission checks completed at journal 21 Apr, 2026 First submitted to journal 20 Apr, 2026 Unknown event 20 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9456965","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":631984549,"identity":"e3e23df5-5afc-47ba-acd1-9b6a8da9d6d2","order_by":0,"name":"Vincent Taschereau-Dumouchel","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0002-9245-7934","institution":"Université de Montréal","correspondingAuthor":true,"prefix":"","firstName":"Vincent","middleName":"","lastName":"Taschereau-Dumouchel","suffix":""},{"id":631984550,"identity":"24980d84-4ccb-4f82-8bc8-a1af2d1f208f","order_by":1,"name":"Marjorie Côté","email":"","orcid":"","institution":"Université de Montréal","correspondingAuthor":false,"prefix":"","firstName":"Marjorie","middleName":"","lastName":"Côté","suffix":""},{"id":631984551,"identity":"95daa249-87c2-4c9f-ab31-3ea47f4aeab7","order_by":2,"name":"Darius Valevicius","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Darius","middleName":"","lastName":"Valevicius","suffix":""},{"id":631984552,"identity":"295a64dc-dfe4-46cf-8acf-71378f3a9172","order_by":3,"name":"Lisa-Marie Davignon","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Lisa-Marie","middleName":"","lastName":"Davignon","suffix":""},{"id":631984553,"identity":"4d205c58-4c2c-4c44-90fd-61c9e32f92bb","order_by":4,"name":"Marie-France Marin","email":"","orcid":"https://orcid.org/0000-0003-0297-5680","institution":"Department of Psychology, Université du Québec à Montréal, Canada","correspondingAuthor":false,"prefix":"","firstName":"Marie-France","middleName":"","lastName":"Marin","suffix":""}],"badges":[],"createdAt":"2026-04-18 16:15:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9456965/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9456965/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108958419,"identity":"e27a1ceb-7aed-48a0-873a-2ab1aa044c7e","added_by":"auto","created_at":"2026-05-11 08:27:05","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":211661,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eExperimental protocols for (A) AFSD, (B) VIFD, and (C) CoFD. \u003c/strong\u003e(A) AFSD (N = 31; 15 women; age 21.5 ± 2.10 years) was acquired in Japan. Participants were selected based on high/very-high fear ratings (6-point Likert scale) for ≥1 of 30 animal categories. Pre-scan instructions emphasized category monitoring. During fMRI, participants monitored blocks of 2, 3, 4, or 6 images (0.98 sec each; fear-relevant animals/neutral animals), pressing buttons to indicate category changes. Only the first images of each block were analyzed across 6 runs (600 per run), for a total of 3600 images presented to the participant. 720 analyzed images per participant. (B) VIFD (N = 67; 34 women; age 23.29 ± 4.21 years) was acquired in Chengdu, China. Instructions emphasized attentive observation and immediate fear ratings. During fMRI, participants viewed three image categories (humans/animals/scenes from IAPS/NAPS/online databases) across 4 runs of 20 pictures. Each trial comprised: (1) 6-sec fixation, (2) 6-sec fear-inducing image, (3) 2-sec fixation, and (4) 4-sec fear rating (5-point Likert scale). 80 images were seen and analyzed per subject. (C) CoFD (N = 153; 118 women, age 26.55 ± 3.12 years) was acquired in Montreal, Canada. The paradigm employed a validated 2-day fear-conditioning/extinction protocol conducted in an fMRI scanner. Three colored lights (yellow, red, and blue) were used as contextual cues. On Day 1, fear acquisition occurred in Context A. Two cues (CS+E and CS+NE) were partially reinforced with electric shocks (5 out of 8 trials), while a third cue (CS−) was never reinforced. Following this, extinction learning occurred in Context B, in which CS+ and CS− were presented without reinforcement. On Day 2, the protocol involved assessing extinction recall with 16 trials of CS+E, CS+NE, and CS- in Context B, followed by fear renewal featuring 4 trials each of CS+E and CS+NE in the original acquisition context (Context A). Each trial consisted of a 12-18 second inter-trial interval, a 3-second context cue, and a 6-second CS presentation (no shocks delivered). Shock expectancy ratings (5-point Likert scale) were collected at the end of each round for the first and last presentation of each CS type. See the original studies for more information. Our analyses focused on extinction, renewal and recall phases to isolate subjective fear representations without confounding shock delivery.\u003ca href=\"https://www.zotero.org/google-docs/?vlUz84\"\u003e\u003csup\u003e5,20,24\u003c/sup\u003e\u003c/a\u003e\u0026nbsp;\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-9456965/v1/b1618953da314bb228bb3f23.png"},{"id":108958448,"identity":"97df265e-7b83-4287-a6c5-132a7b5435e8","added_by":"auto","created_at":"2026-05-11 08:27:07","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":691879,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eWhole-brain neural patterns associated with subjective fear decoding across three independent datasets. \u003c/strong\u003e(A) Combined correlation map showing positive activations (red/yellow) in temporal, occipital, cingulate, and subcortical regions, and negative activations (blue) in prefrontal cortex. (B) Cross-dataset overlap: yellow = all three datasets; green = two datasets; blue = single dataset. Robust consistency observed in anterior cingulate, insula, and lateral occipital cortex. (C) Pairwise comparisons showing highest concordance between AFSD and VIFD.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-9456965/v1/cbaa3acdf0a3019bf2a125b9.png"},{"id":108958408,"identity":"21d31e2f-77c0-426f-9baf-c871a138f791","added_by":"auto","created_at":"2026-05-11 08:27:02","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":484664,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMVPA reveals distributed representations of fear that exhibit limited generalizability, yet some regions remain predictive. \u003c/strong\u003e\u0026nbsp;(A) Predictive performance of the whole-brain decoding models, demonstrating significant Pearson correlations between observed and predicted fear judgments for all datasets (*p ≤ 0.05). (B) Distributed nature of fear representations, showing that decoding accuracy improves with increasing voxel count and plateaus around 10,000 randomly selected voxels. Critically, restricting analysis to a meta-analytic negative affect mask (dashed lines) does not enhance prediction compared to whole-brain voxel selection (solid lines), indicating that fear representations extend beyond traditional emotion-processing regions. (C) Whole-brain activation maps for AFSD, VIFD, and CoFD datasets, showing widespread neural patterns associated with subjective fear ratings across occipital, temporal, parietal, and frontal cortices. (D) Mean prediction correlation (Pearson r) between observed and predicted subjective fear judgments within the Brainnetome Atlas, averaged across the three datasets. Results are FDR-corrected, p \u0026lt; 0.05.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-9456965/v1/66f8d2bb371ec588d1b85113.png"},{"id":108959527,"identity":"feabf07d-8d88-4cdf-a4a1-df70ad0789e6","added_by":"auto","created_at":"2026-05-11 08:30:18","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1939949,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9456965/v1/e6183fbd-d3a4-486a-bbba-24883294a2fd.pdf"},{"id":108958837,"identity":"03852041-8c67-45cd-87a4-a483191225b9","added_by":"auto","created_at":"2026-05-11 08:28:19","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2597988,"visible":true,"origin":"","legend":"Figure S1","description":"","filename":"S1RDM.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9456965/v1/007c4554d53713fc089603e1.pdf"},{"id":108958903,"identity":"8d138989-70af-4194-ba62-3ae4be384e32","added_by":"auto","created_at":"2026-05-11 08:28:27","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":9376011,"visible":true,"origin":"","legend":"Figure S2","description":"","filename":"S2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9456965/v1/16237eed0bc370d6531ef53e.pdf"},{"id":108958440,"identity":"4e3dd2b0-7b35-4e66-8838-015301ede17f","added_by":"auto","created_at":"2026-05-11 08:27:06","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":39881,"visible":true,"origin":"","legend":"Figure S3","description":"","filename":"S3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9456965/v1/fedd0a6ee4e8baa50dcd76ab.pdf"},{"id":108958877,"identity":"32c1613c-0754-4d33-b8aa-447ac117da85","added_by":"auto","created_at":"2026-05-11 08:28:21","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":55640,"visible":true,"origin":"","legend":"Figure S4","description":"","filename":"S4.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9456965/v1/e6f312a7a200934ebd814081.pdf"},{"id":108958404,"identity":"6b48e0f0-9582-4415-a226-295a3b3bfeff","added_by":"auto","created_at":"2026-05-11 08:26:55","extension":"pdf","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":1032733,"visible":true,"origin":"","legend":"Figure S5","description":"","filename":"S5.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9456965/v1/4dbe45e74f585931f857a182.pdf"},{"id":108958421,"identity":"95fcd16a-6903-480a-8323-28615dff59e6","added_by":"auto","created_at":"2026-05-11 08:27:06","extension":"pdf","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":4453669,"visible":true,"origin":"","legend":"Figure S6","description":"","filename":"S6.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9456965/v1/0f9704358e80ca00014d2c7d.pdf"},{"id":108958449,"identity":"c7b12005-6b4d-4c47-87cc-b64667da6061","added_by":"auto","created_at":"2026-05-11 08:27:07","extension":"docx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":3630224,"visible":true,"origin":"","legend":"Supplementary Analysis","description":"","filename":"SupplementaryData.docx","url":"https://assets-eu.researchsquare.com/files/rs-9456965/v1/a9be6743d4bfb1731496e790.docx"}],"financialInterests":"The authors have declared there is \u003cb\u003eNO\u003c/b\u003e conflict of interest to disclose","formattedTitle":"Is the representation of fear distributed across the whole brain?","fulltext":[{"header":"Introduction","content":"\u003cp\u003eFear is a fundamental emotional state that supports survival by enabling organisms to detect, evaluate, and respond to potential threats.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e Despite its central role in adaptive behavior, and its clear implication in anxiety disorders and phobias, there remains substantial debate regarding how subjective fear is represented in the human brain.\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e While decades of neuroimaging research have identified a set of regions consistently engaged during fear-related tasks, the extent to which these regions constitute a coherent and generalizable neural representation of subjective fear remains unresolved.\u003csup\u003e\u003cspan additionalcitationids=\"CR4 CR5 CR6 CR7\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eMeta-analytic and mass-univariate studies have converged on a so-called canonical fear network, typically encompassing the amygdala, anterior cingulate cortex, insula, hippocampus, and prefrontal regions.\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan additionalcitationids=\"CR10 CR11 CR12 CR13 CR14\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e This network overlaps considerably with broader systems involved in negative affect and salience processing, suggesting that fear may not rely on a strictly specialized neural substrate. Yet, these approaches primarily emphasize localized activations and may underestimate the contribution of distributed neural signals that jointly encode subjective experience.\u003csup\u003e\u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eRecent advances in multivariate pattern analysis (MVPA) have challenged this canonical view by modelling fear as a distributed pattern of activity spanning multiple brain regions simultaneously.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e Using MVPA, a previous study reported that subjective fear is better predicted by activity in prefrontal, occipital, and ventral temporal cortices, whereas physiological threat responses are more closely linked to subcortical and interoceptive regions such as the amygdala and insula.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e These findings have been interpreted as evidence for a dissociation between neural systems supporting threat detection and those supporting the conscious experience of fear.\u003c/p\u003e \u003cp\u003eHowever, this interpretation has been challenged by Zhou and colleagues, who reported that subjective fear could be predicted equally well using randomly sampled voxels distributed across the entire brain as by any specific functional network.\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e On this basis, the authors argued that subjective fear may lack a privileged neural locus and instead emerge from highly distributed brain-wide representations. This conclusion is rather surprising as their own mass-univariate analyses revealed a network that substantially overlapped with the canonical fear system (Sup. Figure\u0026nbsp;3a in Zhou et al., 2021). This apparent discrepancy raises a critical question: do MVPA results indicate that the notion of a canonical fear network is fundamentally wrong, or do they rather reflect differences in how neural information is captured and evaluated by those two different statistical approaches?\u003c/p\u003e \u003cp\u003eOne possibility is that MVPA can leverage subtle activation patterns distributed across many brain regions, including those outside traditional emotion networks.\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e While these patterns may improve prediction, they might reflect correlated processes (e.g., attention, arousal) rather than fear representation per se.\u003csup\u003e21\u003c/sup\u003e Another possibility is that information captured outside the canonical network reflects task-specific or dataset-specific features that correlate with fear ratings but do not generalize across paradigms. Another consideration pertains to the studied networks. Zhou et al.\u0026rsquo;s analyses tested predefined resting-state and \u0026ldquo;consciousness\u0026rdquo; networks but not the canonical fear network itself, leaving open the question of whether fear-related information is meaningfully concentrated within regions previously identified by mass-univariate meta-analyses.\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eClarifying this issue is particularly important for translational research. If subjective fear is encoded in a stable and generalizable neural network, this network may represent a viable target for clinical interventions. Conversely, if fear-related information is highly distributed and largely context-dependent, neural markers derived from MVPA may show limited generalizability despite strong within-sample performance.\u003c/p\u003e \u003cp\u003eIn the present study, we directly address this issue by combining mass-univariate and multivariate approaches across three independent fMRI datasets capturing distinct forms of subjective fear: fear of personally relevant animals, fear of diverse threatening images, and fear of conditioned threat cues. By examining whole-brain decoding performance, voxel sampling strategies, and region-wise predictive capacity, we aim to determine whether subjective fear is primarily encoded within a canonical neural network, and whether distributed information outside this network reflects core, generalizable representations or paradigm-specific information.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eParticipants and experimental design\u003c/h2\u003e \u003cp\u003eWe leverage three functional magnetic resonance imaging (fMRI) datasets that collected participants' subjective fear ratings: (1) the Animal Fear Schema Dataset (AFSD; n\u0026thinsp;=\u0026thinsp;31), in which participants with self-reported elevated fear of specific animals were presented with a series of images of animals; (2) the Visually Induced Fear Dataset (VIFD; n\u0026thinsp;=\u0026thinsp;67), which includes the presentation of diverse threatening stimuli (scenes, animals, objects) presented to healthy volunteers screened for psychiatric disorders; and (3) the Conditioned Fear Dataset (CoFD; n\u0026thinsp;=\u0026thinsp;153), which includes healthy volunteers screened for psychiatric/medical disorders undergoing a classical fear conditioning paradigm, where participants were invited to provide their subjective report of shock likelihood.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eFor CoFD, we analyzed extinction, renewal and recall phases rather than initial acquisition to avoid confounding subjective fear with pain responses to shock delivery. During extinction (Day 1) and renewal (Day 2), no shocks were administered, allowing us to isolate learned fear representations independent of nociceptive processing. This approach ensures that shock expectancy ratings reflect anticipated threat rather than responses to actual aversive stimuli. See Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e for more details on experimental protocols.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eData preprocessing\u003c/h3\u003e\n\u003cp\u003ePreprocessing for AFSD and VIFD is described in the original publications.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e For CoFD, first-level GLMs were computed using SPM12 with motion regressors and high-pass filtering (128s cutoff). We excluded 22 participants due to missing data, technical errors, or generalized fear responses (high fear for CS-). For each dataset, fear ratings were combined with neuroimaging data and averaged within-subject across fear levels, yielding beta files for up to six subjective fear levels per participant. See supplementary data for more information. Datasets were spatially aligned and mean-centered within participants.\u003c/p\u003e\n\u003ch3\u003eAnalyses\u003c/h3\u003e\n\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eMass-univariate analyses\u003c/h2\u003e \u003cp\u003eMass univariate analyses are the typical approach for analyzing localized neural activations in relation to an experimental stimulus.\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e In contrast to MVPA, mass univariate analyses assess the relationship between each voxel and the experimental variable independently, typically followed by multiple-comparison correction, such as false discovery rate (FDR) control.\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e Usually, this takes the form of a generalized linear model (GLM), which estimates betas (slopes) and t-statistics between the time series of each voxel in the brain and a representation of the experimental variable which has been convolved with a canonical haemodynamic response function (HRF), while accounting for other sources of noise such as scanner drift, motion, and physiological noise.\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eHowever, for two of the three studies, data were already reduced to a single beta map per fear level and per participant. As we did not have access to the time-series data itself, we opted for a simplified mass-univariate analysis, pooling subject data within studies and fitting multilevel regression models (MLMs) for each voxel, with subject ID as a grouping factor and a random intercept. The independent fixed-effects variable was the voxel's beta value, and the dependent variable was fear level. This produced one spatial map of beta coefficients and p-values for each of the three datasets, which were then combined using Fisher's z-transform method. The p-values were then adjusted for multiple comparisons using an FDR correction. The results are presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e (with adjusted p-values averaged by Brainnetome ROIs) and Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA. Additionally, we computed Pearson correlations to visualize cross-dataset agreement (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB-C).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cb\u003eBrain regions showing significant associations with subjective fear ratings in combined mass-univariate analysis (AFSD, VIFD, CoFD; N\u0026thinsp;=\u0026thinsp;251).\u003c/b\u003e Anatomical labels from Brainnetome Atlas; β-values from combined MLM analysis; p-values combined via Fisher's z-transform. See text for details.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLobe\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGyrus\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRegion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eHemisphere\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eb-value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFDR-corrected p-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"8\" rowspan=\"9\"\u003e \u003cp\u003eFrontal lobe\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eSFG, Superior Frontal Gyrus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA8m, medial area 8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA6dl, dorsolateral area 6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.047\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMFG, Middle Frontal Gyrus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA10l, lateral area10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.025\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003eIFG, Inferior Frontal Gyrus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA44d,dorsal area 44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.043\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA44d,dorsal area 44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA45c, caudal area 45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA44op, opercular area 44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA44op, opercular area 44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePrG, Precentral Gyrus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA6cvl, caudal ventrolateral area 6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.044\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"12\" rowspan=\"13\"\u003e \u003cp\u003eTemporal Lobe\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMFG, Middle Temporal Gyrus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA37dl, dorsolateral area37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.024\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eITG, Inferior Temporal Gyrus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA37elv, extreme lateroventral area37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA37vl, ventrolateral area 37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.025\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA37vl, ventrolateral area 37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eFuG, Fusiform Gyrus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA37mv, medioventral area37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA37mv, medioventral area37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA37lv, lateroventral area37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.015\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA37lv, lateroventral area37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003ePhG, Parahippocampal gyrus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA35/36r, rostral area 35/36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA35/36r, rostral area 35/36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.043\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTI, area TI(temporal agranular insular cortex)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.012\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003epSTS, Posterior Superior Temporal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ecpSTS, caudoposterior superior temporal sulcus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.007\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ecpSTS, caudoposterior superior temporal sulcus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.023\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"6\" rowspan=\"7\"\u003e \u003cp\u003eParietal Lobe\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"5\" rowspan=\"6\"\u003e \u003cp\u003eSPL, Superior Parietal Lobule\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA7r, rostral area 7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA7c, caudal area 7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA5l, lateral area 5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.016\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA5l, lateral area 5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.046\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA7ip, intraparietal area 7(hIP3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.009\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA7ip, intraparietal area 7(hIP3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.027\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIPL, Inferior Parietal Lobule\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA40rd, rostrodorsal area 40(PFt)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.031\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"5\" rowspan=\"6\"\u003e \u003cp\u003eInsular Lobe\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"5\" rowspan=\"6\"\u003e \u003cp\u003eINS, Insular Gyrus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003evIa, ventral agranular insular\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003evIa, ventral agranular insular\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.009\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003edIa, dorsal agranular insular\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003edIa, dorsal agranular insular\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003edId, dorsal dysgranular insular\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.016\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003edId, dorsal dysgranular insular\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.044\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"5\" rowspan=\"6\"\u003e \u003cp\u003eLimbic Lobe\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"5\" rowspan=\"6\"\u003e \u003cp\u003eCG, Cingulate Gyrus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA24rv, rostroventral area 24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.025\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA32p, pregenual area 32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.008\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA32p, pregenual area 32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.028\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA24cd, caudodorsal area 24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA24cd, caudodorsal area 24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA23c, caudal area 23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.033\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"5\" rowspan=\"6\"\u003e \u003cp\u003eOccipital Lobe\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"5\" rowspan=\"6\"\u003e \u003cp\u003eLOcC, Lateral Occipital Cortex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003emOccG, middle occipital gyrus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.028\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eV5/MT+, area V5/MT+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eV5/MT+, area V5/MT+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eiOccG, inferior occipital gyrus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.034\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eiOccG, inferior occipital gyrus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003elsOccG, lateral superior occipital gyrus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"13\" rowspan=\"14\"\u003e \u003cp\u003eSubcortical Nuclei\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eAmyg, Amygdala\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003emAmyg, medial amygdala\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003emAmyg, medial amygdala\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.014\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003elAmyg, lateral amygdala\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.009\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003elAmyg, lateral amygdala\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.015\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eBG, Basal Ganglia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGP, globus pallidus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.031\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003evmPu, ventromedial putamen\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.015\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003evmPu, ventromedial putamen\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.032\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"6\" rowspan=\"7\"\u003e \u003cp\u003eTha, Thalamus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003emPFtha, medial pre-frontal thalamus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003emPFtha, medial pre-frontal thalamus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003emPMtha, pre-motor thalamus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.008\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eStha, sensory thalamus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePPtha, posterior parietal thalamus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.038\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003elPFtha, lateral pre-frontal thalamus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.015\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003elPFtha, lateral pre-frontal thalamus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.008\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eWhole-brain decoding performance across datasets\u003c/h3\u003e\n\u003cp\u003eTo predict subjective fear ratings from whole-brain activation patterns, we applied Support Vector Regression (SVR; CANlab toolbox, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/canlab/CanlabCore\u003c/span\u003e\u003cspan address=\"https://github.com/canlab/CanlabCore\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) with leave-one-subject-out cross-validation to each dataset and to the combined datasets. Model performance was evaluated using Pearson correlations between observed and predicted fear ratings, and area under the ROC curve (AUC) for discriminating low-fear (0\u0026ndash;2) from high-fear (3\u0026ndash;5) trials derived from continuous SVR outputs. Statistical significance was assessed via 1,000 permutations. To examine cross-dataset transferability, models trained on each dataset were applied to the remaining two datasets, with performance evaluated using both metrics and FDR-corrected p-values across all cross-prediction tests.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eNegative Affect Mask Decoder\u003c/h2\u003e \u003cp\u003eTo test Zhou et al.'s proposition that fear lacks a unique neural locus, we implemented a voxel-sampling approach examining how prediction accuracy scales with brain coverage.\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e We trained SVR models on: 1) randomly selected voxel subsets (100 to 348,904 voxels; 15 iterations per subset), and 2) voxels constrained to Lindquist et al.'s meta-analytic \"negative affect\" mask available on Neurosynth (negative affect_uniformity-test_z_FDR.01.nii).\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e This allowed direct comparison of decoder performance within the canonical network versus random, spatially distributed voxels. Performance was measured using Pearson correlations between observed and predicted fear ratings, with FDR correction across six model comparisons (3 datasets \u0026times; 2 conditions: whole-brain vs. mask).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eROI Analysis\u003c/h3\u003e\n\u003cp\u003eNext, we analyzed the predictive capacity of the decoders from each dataset to predict subjective reports across 214 regions of the Brainnetome Atlas, a brain parcellation atlas based on brain connectivity, including 210 cortical regions and four subcortical regions (bilateral amygdala and hippocampi).\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e For each Brainnetome ROI, we extracted the voxel data by applying the corresponding mask and predicted the subjective ratings using a Support Vector Regression (SVR) model in a leave-one-subject-out cross-validation procedure. Model performance was assessed by computing Pearson correlations between real and predicted fear ratings. To statistically compare correlation coefficients across datasets, we applied Fisher's z-transformation, which converts correlation coefficients into z-scores, enabling statistical comparison between datasets.\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e We assessed the significance of correlation coefficients within each dataset by correcting for multiple comparisons across ROIs using the False Discovery Rate (FDR) procedure (Benjamini-Hochberg, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, assuming dependent tests).\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eMass-univariate analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCombining the three datasets (AFSD, VIFD, and CoFD) revealed significant positive activations (red and yellow) in the prefrontal cortex (superior and inferior frontal gyrus), the precentral gyrus, the postcentral gyrus, the anterior cingulate and insular gyrus, and subcortical structures including the amygdala, basal ganglia and thalamus (Fig. 2A and Table 1). Additional positive activations were observed in the temporal (inferior, middle, fusiform, parahippocampal and posterior temporal gyrus), parietal (superior and inferior lobules) and lateral occipitotemporal cortex. Negative activations (blue) were primarily located in the middle frontal gyrus (Table 1). Overall, these results indicate that fear is represented in a network including the \u0026nbsp;prefrontal, temporal, occipital, cingulate, and subcortical regions, alongside decreased activity in medial prefrontal areas. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe overlap map highlights the degree of consistency across datasets (Fig. 2B). Most voxels showing significant activations in all three datasets were located in the anterior cingulate gyrus, insula, \u0026nbsp; diencephalon, mesencephalon, genu of the corpus callosum, and lateral occipitotemporal cortex (yellow clusters). Areas unique to a single dataset (blue) were primarily confined to occipital and inferior temporal cortices, suggesting potential task- or stimulus-related variability. Peak coordinates and statistical values for all significant clusters are reported in Table 1.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWhole-brain decoding performance across datasets\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe first examined whether subjective fear ratings could be predicted from whole-brain activation patterns across all three datasets. Support Vector Regression models with leave-one-subject-out cross-validation revealed significant above-chance prediction accuracy for all datasets (Fig. 3 and Table 2). The AFSD showed the strongest decoding performance (r = 0.749, p \u0026lt; 0.001; AUC = 0.880, p = 0.002), followed by VIFD (r = 0.574, p \u0026lt; 0.001; AUC = 0.795, p = 0.002) and CoFD (r = 0.447, p \u0026lt; 0.001; AUC = 0.750, p = 0.002). Whole-brain activation maps revealed that subjective fear was associated with distributed neural responses across the cortex in all three datasets (see Fig. 3C). Rather than being restricted to specific emotion-related regions, activation patterns were widespread, involving occipital, temporal, parietal, and frontal areas.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2: Correlation of subjective fear judgments with predicted fear from brain activity according to each dataset\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"619\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 129px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDataset\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePearson Correlation (r)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ecorrected p-value (\u003cem\u003ep\u003c/em\u003e)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eArea Under the Curve (AUC)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 143px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ecorrected p-value (\u003cem\u003ep\u003c/em\u003e)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 129px;\"\u003e\n \u003cp\u003eAFSD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e0.749*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0.880*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 143px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 129px;\"\u003e\n \u003cp\u003eVIFD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e0.574*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0.795*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 143px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 129px;\"\u003e\n \u003cp\u003eCoFD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e0.447*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0.750*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 143px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 129px;\"\u003e\n \u003cp\u003eAFSD(\u003cem\u003eVIFD\u003c/em\u003e)\u003cstrong\u003e\u003csup\u003e1\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e0.627*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0.803*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 143px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 129px;\"\u003e\n \u003cp\u003eAFSD(\u003cem\u003eCoFD\u003c/em\u003e)\u003cstrong\u003e\u003csup\u003e1\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e0.285*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0.601*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 143px;\"\u003e\n \u003cp\u003e0.009\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 129px;\"\u003e\n \u003cp\u003eVIFD(\u003cem\u003eAFSD\u003c/em\u003e)\u003cstrong\u003e\u003csup\u003e1\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e0.182*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0.608*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 143px;\"\u003e\n \u003cp\u003e0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 129px;\"\u003e\n \u003cp\u003eVIFD(\u003cem\u003eCoFD\u003c/em\u003e)\u003cstrong\u003e\u003csup\u003e1\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e0.191*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0.581*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 143px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 129px;\"\u003e\n \u003cp\u003eCoFD(\u003cem\u003eAFSD\u003c/em\u003e)\u003cstrong\u003e\u003csup\u003e1\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e0.098*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e0.025\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0.565*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 143px;\"\u003e\n \u003cp\u003e0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 129px;\"\u003e\n \u003cp\u003eCoFD(\u003cem\u003eVIFD\u003c/em\u003e)\u003cstrong\u003e\u003csup\u003e1\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e0.190*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0.633*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 143px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003csup\u003e1\u003c/sup\u003e Cross-prediction to see if brain patterns learned in one model dataset can be generalized to another dataset where \u0026ldquo;Data(\u003cem\u003eModel\u003c/em\u003e)\u0026rdquo;. * = p \u0026lt; 0.05 after FDR correction.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCross-dataset generalization analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo assess whether fear representations generalized across datasets, we applied decoders trained on one dataset to predict fear ratings in the others (Table 2). Generally, models showed modest generalizability to other datasets. Models trained on VIFD showed strong generalization to AFSD (r = 0.627, p \u0026lt; 0.001; AUC = 0.803, p = 0.002) and weak generalization to CoFD (r = 0.190, p \u0026lt; 0.001; AUC = 0.633, p = 0.002). Conversely, models trained on AFSD generalized weakly to VIFD (r = 0.182, p \u0026lt; 0.001; AUC = 0.608, p = 0.003) and to CoFD (r = 0.098, p = 0.025; AUC = 0.565, p = 0.003). Models trained on CoFD similarly showed weak generalization to both AFSD (r = 0.285, p \u0026lt; 0.001; AUC = 0.601, p = 0.003) and VIFD (r = 0.191, p \u0026lt; 0.001; AUC = 0.591, p = 0.002).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRestricting predictions to the canonical fear mask\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFigure 3B shows Pearson correlations between observed and predicted fear ratings across all three datasets as the number of randomly selected voxels increased from 100 to 348,904 voxels. When we restricted analyses to voxels within the Lindquist meta-analytic mask, we observed distinct patterns across datasets. In VIFD and COFD, constraining predictions to the canonical emotion network decreased performance relative to random voxel selection, suggesting that the decoders likely rely on information outside these regions to make predictions. In contrast, for AFSD, prediction accuracy within the mask closely matched that of random selection, suggesting that little unique information is represented outside the canonical network in this dataset.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePrediction accuracy using all brain voxels was significant for all three datasets (see Table 2 or Fig. 3A for detailed statistics). When restricting analyses to the canonical fear mask, performance remained significant and showed similar patterns (AFSD: r = 0.681, \u0026nbsp;p \u0026lt; 0.001; VIFD: r = 0.358, \u0026nbsp;p \u0026lt; 0.001; CoFD: r = 0.274, \u0026nbsp;p \u0026lt; 0.001). At approximately 10,000 randomly selected voxels, performance reached a plateau, and mean correlations were 0.746 \u0026plusmn; 0.008 (SD) for AFSD, 0.556 \u0026plusmn; 0.013 for VIFD, and 0.426 \u0026plusmn; 0.021 for CoFD. When restricting analyses to the canonical fear mask, performance also plateaued at approximately 10,000 randomly selected voxels, and mean correlations were 0.681 \u0026plusmn; 0.009 (SD) for AFSD, 0.355 \u0026plusmn; 0.010 for VIFD, and 0.264 \u0026plusmn; 0.014 for CoFD.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe decoder trained on the three combined datasets failed to reach statistical significance (p \u0026gt; 0.05), indicating no reliable predictive performance.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCommon regions for the prediction of fear across datasets\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRegion-by-region analyses across 214 brain regions from the Brainnetome Atlas identified 23 common regions that showed significant brain-subjective fear correlations across all three datasets (Pearson r = 0.182\u0026ndash;0.306; FDR-corrected p \u0026lt; 0.05 within each dataset; Figure 3D, Table 3). These regions include anterior cingulate gyrus, insular cortex (agranular insular), visual processing areas (lateral and medioventral occipital cortex), memory-related structures (parahippocampal gyrus), temporal cortex (fusiform and inferior temporal gyri), orbitofrontal cortex (orbital gyrus), parietal regions (precuneus), sensorimotor areas (paracentral lobule), prefrontal cortex (superior and inferior frontal gyri), and subcortical nuclei (amygdala). The strongest mean correlations were observed in left area V5/MT+ of the lateral occipital cortex (r = 0.306), right medioventral area 37 of the fusiform gyrus (r = 0.300), and left pregenual area 32 of the cingulate gyrus (r = 0.286).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 3. Regions with significant brain-behaviour correlations across all datasets (N = 251; FDR-corrected q \u0026lt; 0.05).\u003c/strong\u003e SVR predictions with leave-one- subject-out cross-validation. Mean r and p-values across datasets shown.\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"617\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLobe\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGyrus\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRegion\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHemisphere\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e\u003cstrong\u003er-value (mean)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFDR-corrected p-value (mean)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"7\" valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003eFrontal lobe\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eSFG, Superior Frontal Gyrus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eA8m, medial area 8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eRight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.204\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eA9l, lateral area 9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eRight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.212\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eIFG, Inferior Frontal Gyrus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eA45r, rostral area 45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eLeft\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.200\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eA44v, ventral area 44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eLeft\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.182\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eOrG, Orbital Gyrus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eA11l, lateral area 11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eRight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.269\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eA12/47l, lateral area 12/47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eRight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.270\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003ePCL, Paracentral Lobule\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eA1/2/3ll, area 1/2/3 (lower limb)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eRight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.213\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"5\" valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003eTemporal Lobe\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eITG, Inferior Temporal Gyrus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eA20cv, caudoventral area 20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eLeft\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.252\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eFuG, Fusiform Gyrus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eA37mv, medioventral area37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eRight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.300\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eA37lv, lateroventral area37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eRight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.264\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003ePhG, Parahippocampal gyrus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eTI, area TI\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eLeft\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.285\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eTH, area TH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eRight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.278\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\" valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003eParietal Lobe\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"3\" valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003ePcu, Precuneus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eA5m, medial area 5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eLeft\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.264\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eA5m, medial area 5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eRight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.226\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eA31m, medial area 31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eRight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.217\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003eInsular Lobe\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eINS, Insular Gyrus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003evIa, ventral agranular insular\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eLeft\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.225\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003edIa, dorsal agranular insular\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eRight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.243\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003eLimbic Lobe\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eCG, Cingulate Gyrus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eA32p, pregenual area 32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eLeft\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.286\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003eOccipital Lobe\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eMVOcC, Medioventral Occipital Cortex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003erLingG, rostral lingual gyrus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eRight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.282\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003evmPOS, ventromedial parietooccipital sulcus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eRight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.270\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eLOcC, Lateral Occipital Cortex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eV5/MT+, area V5/MT+\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eLeft\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.306\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003eiOccG, inferior occipital gyrus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eLeft\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.247\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 60px;\"\u003e\n \u003cp\u003eSubcortical Nuclei\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eAmyg, Amygdala\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003elAmyg, lateral amygdala\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 92px;\"\u003e\n \u003cp\u003eLeft\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e0.212\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 116px;\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eMass-univariate analyses and MVPA combined\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWhen comparing our two primary analyses (mass-univariate and MVPA), we found eight regions that were consistently identified across all three datasets: the fusiform gyrus (right medioventral and lateroventral area 37; b = 0.1 and b = 0.12; r\u003csub\u003emean\u003c/sub\u003e= 0.300 and r\u003csub\u003emean\u003c/sub\u003e= 0.264), the insula (left ventral agranular and right dorsal dysgranular; b = 0.1 and b = 0.12; r\u003csub\u003emean\u003c/sub\u003e= 0.225 and r\u003csub\u003emean\u003c/sub\u003e= 0.243), the cingulate gyrus (right pregenual area 32; b = 0.1; r\u003csub\u003emean\u003c/sub\u003e= 0.286), the occipital gyrus (left V5/MT+ and left inferior occipital gyrus; b = 0.12 and b = 0.07; r\u003csub\u003emean\u003c/sub\u003e= 0.306 and r\u003csub\u003emean\u003c/sub\u003e= 0.247), and the left lateral amygdala (b = 0.08; r\u003csub\u003emean\u003c/sub\u003e= 0.212) (see Table 1 and 3).\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study addressed an important question in affective neuroscience: Is subjective fear represented in a localized canonical network or distributed across the entire brain?\u003csup\u003e20\u003c/sup\u003e By combining mass-univariate and multivariate pattern analyses across three independent fMRI datasets (N\u0026thinsp;=\u0026thinsp;251 total) capturing distinct fear contexts (animal-specific fear, visually-induced situational fear, and conditioned fear cues), we provide evidence for both perspectives.\u003c/p\u003e \u003cp\u003eOur key findings are threefold. First, mass-univariate analyses and MVPA identified a consistent network spanning the anterior cingulate cortex, insula, fusiform gyrus, lateral occipital cortex (V5/MT+), and amygdala. Second, whole-brain MVPA demonstrated that distributed patterns significantly predict subjective fear, consistent with Zhou et al.'s findings. Third, critically, restricting analyses to a meta-analytic negative affect mask did not improve prediction over random voxel selection, yet the eight convergent regions exhibited both robust focal activation and informative multivariate patterns. These findings reconcile apparently contradictory perspectives by demonstrating that a canonical network exists but does not contain all fear-relevant information, and that distributed information outside this network appears to be primarily paradigm-specific rather than generalizable.\u003c/p\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eThe canonical network of fear: functional specialization and integration\u003c/h2\u003e \u003cp\u003eThe eight consistently identified regions from mass-univariate and MVPA approaches form a functionally coherent network integrating threat detection, interoceptive awareness, visual processing, and conscious emotional experience.\u003c/p\u003e \u003cp\u003e \u003cb\u003eAmygdala and threat detection.\u003c/b\u003e The presence of the left lateral amygdala is consistent with its established role in threat detection and associative learning, with recent evidence suggesting the left amygdala specifically mediates cognitively-mediated fear acquisition and extinction.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e Its recruitment across fear conditions underscores its function as a rapid and reliable threat detector, operating early in the fear processing hierarchy to initiate downstream cortical and subcortical responses.\u003c/p\u003e \u003cp\u003e \u003cb\u003eInsula and interoceptive integration\u003c/b\u003e. Bilateral insula activation likely reflects the integration of bodily signals (heart rate, muscle tension, visceral sensations) \u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e The insula's role as an interoceptive hub makes it well-positioned to transform physiological arousal into subjective feeling states. Its consistent engagement across paradigms suggests that subjective fear requires awareness of bodily changes, regardless of whether the trigger is a phobic animal, threatening scene, or conditioned stimulus.\u003c/p\u003e \u003cp\u003e \u003cb\u003eAnterior cingulate cortex and emotional expression\u003c/b\u003e. The pregenual ACC's involvement aligns with its documented role in cognitive control, conflict monitoring, and emotion-related appraisal, including fear evaluation.\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e This region may adjudicate between competing interpretations of ambiguous stimuli and modulate the intensity of fear expression. Its activation could reflect the evaluative component of fear: \"How threatening is this stimulus to me?\"\u003c/p\u003e \u003cp\u003e \u003cb\u003eFusiform gyrus as a visual-emotional interface.\u003c/b\u003e The fusiform gyrus is broadly involved in high-level visual processing, including fine-grained object discrimination, perceptual expertise, and category-sensitive representations.\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e Through its connectivity with the anterior temporal lobe, which supports semantic memory, and with medial temporal structures such as the hippocampus, the fusiform gyrus is well positioned to integrate visual representations with mnemonic and emotional context.\u003csup\u003e\u003cspan additionalcitationids=\"CR36\" citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e Fearful stimuli, whether animals, threatening scenes, or conditioned cues, may amplify perceptual processing within ventral occipito-temporal cortex when they are motivationally relevant, a process that likely draws on these multimodal connections.\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e Importantly, a multivoxel decoding study suggests that ventral occipito-temporal regions contribute significantly to predicting subjective fear ratings, and in some cases, fusiform activity patterns predict self-reported fear more strongly than peripheral autonomic measures, suggesting that these representations are more closely linked to conscious experience than to purely defensive responses.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e In our data, the right medioventral and lateroventral fusiform (area 37) exhibited among the strongest and most consistent cross-dataset associations in both mass-univariate and ROI decoding analyses, reinforcing this interpretation. This interpretation is consistent with evidence suggesting that motivationally relevant stimuli, including fear-inducing ones, may recruit high-level visual mechanisms.\u003csup\u003e\u003cspan additionalcitationids=\"CR40 CR41\" citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e When stimuli are fear-inducing, their visual features may thus be encoded with greater resolution or gain, increasing their subjective salience. This is consistent with evidence showing that emotional salience and attention jointly modulate fusiform responses, and that fusiform\u0026ndash;amygdala connectivity increases in anxiety states.\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e The fusiform gyrus may therefore act as a visual\u0026ndash;emotional amplification hub, where perceptual representations become tightly coupled with affective evaluation.\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003e \u003cb\u003eLateral occipital cortex and threat feature extraction.\u003c/b\u003e The lateral occipital cortex, encompassing area V5/MT\u0026thinsp;+\u0026thinsp;and the inferior occipital gyrus, is classically associated with motion processing, visual feature extraction, and the perception of biological movement.\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e,\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e Notably, the engagement of V5/MT+ across all three paradigms, including conditioned light cues devoid of any inherent motion, suggests that its activation is unlikely to reflect low-level stimulus features alone. Instead, it may reflect top-down modulation driven by the learned or intrinsic motivational relevance of the stimulus, consistent with evidence that the amygdala can amplify activity in motion-sensitive visual regions in the absence of actual movement.\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e Beyond motion per se, the inferior occipital gyrus likely contributes lower-level feature extraction that feeds into the higher-level representations computed in fusiform and parahippocampal regions.\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e Together, these lateral occipital regions may constitute an early visual gateway for threat prioritization, rapidly extracting stimulus features that are subsequently elaborated by downstream temporal and limbic structures.\u003csup\u003e\u003cspan additionalcitationids=\"CR50\" citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e Consistent with this view, prior work demonstrated that the ventral visual stream encodes affectively neutral visual features statistically associated with fear ratings, independently of whether subjective fear is actually reported, and that it is the multivariate information transmission between ventral visual areas and prefrontal regions that distinguishes participants reporting subjective fear from those who do not.\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eDistributed representations: separating signal from noise\u003c/h2\u003e \u003cp\u003eOur voxel-sampling analyses replicate Zhou et al.'s finding that prediction improves with increasing brain coverage, seemingly supporting the view that subjective fear lacks a privileged neural locus. However, we made three critical observations: First, cross-dataset generalization was weak (r\u0026thinsp;\u0026lt;\u0026thinsp;0.30 for most comparisons), indicating that whole-brain decoders primarily capture paradigm-specific variance rather than generalizable fear representations. A truly distributed, content-general fear representation should transfer robustly across contexts and our results suggest otherwise. Second, restricting analyses to a meta-analytic negative affect network did not enhance prediction in VIFD and CoFD, confirming that decoders leverage information outside canonical emotion regions. Only AFSD showed comparable performance within versus outside this network, possibly reflecting stronger canonical circuit engagement in consolidated fear schemas relative to situational or conditioned fear. Third, the combined-dataset decoder failed to reach significance (p\u0026thinsp;\u0026gt;\u0026thinsp;0.05): if fear were truly represented in a universal distributed manner, pooling datasets should enhance power, but instead obscures shared canonical signals, suggesting that much of the distributed signal may reflect idiosyncratic task demands rather than core fear processes. Collectively, these findings suggest a hybrid model: a canonical network reliably represents core fear components across contexts, while distributed patterns may capture context-specific elaborations that improve within-sample prediction but generalize poorly across datasets.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eReconciling seemingly contradictory approaches\u003c/h2\u003e \u003cp\u003eThe apparent contradiction between mass-univariate and MVPA findings is best understood as a reflection of each approach's design rather than a fundamental disagreement about the neural basis of fear. Mass-univariate analyses detect locally robust, spatially focal activations that replicate across paradigms, whereas MVPA exploits multivariate spatial patterns that may improve within-sample prediction by leveraging task-specific covariance structures.\u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e,\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e These approaches therefore operate on different sources of neural variance, and their convergence, rather than their divergence, is most informative. The eight regions identified by both methods provide the strongest evidence for a canonical network, exhibiting both robust focal activation and informative patterns across all three datasets. Critically, the failure of the combined-dataset decoder (p\u0026thinsp;\u0026gt;\u0026thinsp;0.05) further supports this interpretation: pooling datasets without accounting for task-specific variance obscures shared canonical signals, underscoring that the generalizable component of fear-related neural activity is concentrated within a reproducible set of regions rather than diffusely distributed across the brain.\u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eMethodological considerations and limitations\u003c/h2\u003e \u003cp\u003eSeveral limitations should be noted. Gender was anonymized in VIFD, limiting our ability to examine sex differences in fear processing.\u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e,\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e Experimental designs differed substantially across datasets in presentation duration, rating timing, trial numbers (720 vs. 80 vs. 16 analyzed trials), and rating scales. These differences may explain varying decoding performance and limited cross-dataset generalization. Furthermore, we relied on a single dataset per paradigm type, limiting our ability to distinguish generalizable from dataset-specific features. Future work should replicate findings across multiple datasets per paradigm type or employ within-subject designs that combine multiple fear paradigms (e.g., specific, general, and conditioned fear).\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eOur findings support the hypothesis of a canonical fear network comprising the fusiform gyrus, insula, anterior cingulate cortex, lateral occipital cortex, and amygdala. The convergence of mass-univariate and MVPA approaches demonstrates that methodological choices are critical. Mass-univariate analyses may be tailored to identify robust, generalizable focal effects, while MVPA may capture additional information that may be task-specific. Considering these methodological issues is essential for translational neuroscience seeking to identify neural predictors of treatment response. These findings have important clinical implications for anxiety disorders and phobias. The canonical fear network identified here provides reliable neural targets for therapeutic intervention, where intervention targeting the modulation of these regions might yield consistent therapeutic effects. Future studies combining multiple fear paradigms within participants are recommended to distinguish methodological from experiential variance.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eConflicts of interest\u003c/h2\u003e \u003cp\u003eAuthors declare that there are not any competing financial interests in relation to the work described.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eMarjorie C\u0026ocirc;t\u0026eacute; was supported by the Natural Sciences and Engineering Research Council of Canada and Dr Vincent Taschereau-Dumouchel was supported in part by the Fond de recherche du Qu\u0026eacute;bec - Sant\u0026eacute; and the Fondation de l\u0026rsquo;Institut universitaire en sant\u0026eacute; mentale de Montr\u0026eacute;al.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eMobbs, D. The ethological deconstruction of fear(s). \u003cem\u003eCurr. Opin. Behav. Sci.\u003c/em\u003e 24, 32\u0026ndash;37 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarcia, R. Neurobiology of fear and specific phobias. \u003cem\u003eLearn. Mem.\u003c/em\u003e 24, 462\u0026ndash;471 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeDoux, J. \u0026amp; Brown, R. A higher-order theory of emotional consciousness. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 114, E2016\u0026ndash;E2025 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeDoux, J. \u0026amp; Pine, D. S. Using Neuroscience to Help Understand Fear and Anxiety: A Two-System Framework. \u003cem\u003eAm. J. Psychiatry\u003c/em\u003e 173, 1083\u0026ndash;1093 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaschereau-Dumouchel, V., Kawato, M. \u0026amp; Lau, H. Multivoxel pattern analysis reveals dissociations between subjective fear and its physiological correlates. \u003cem\u003eMol. Psychiatry\u003c/em\u003e 25, 2342\u0026ndash;2354 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaschereau-Dumouchel, V., Michel, M., Lau, H., Hofmann, S. G. \u0026amp; LeDoux, J. E. Putting the \u0026lsquo;mental\u0026rsquo; back in \u0026lsquo;mental disorders\u0026rsquo;: a perspective from research on fear and anxiety. \u003cem\u003eMol. Psychiatry\u003c/em\u003e 27, 1322\u0026ndash;1330 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWendt, J., L\u0026ouml;w, A., Weymar, M., Lotze, M. \u0026amp; Hamm, A. O. Active avoidance and attentive freezing in the face of approaching threat. \u003cem\u003eNeuroImage\u003c/em\u003e 158, 196\u0026ndash;204 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMauss, I. B. \u0026amp; Robinson, M. D. Measures of emotion: A review. \u003cem\u003eCogn. Emot.\u003c/em\u003e 23, 209\u0026ndash;237 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGilmartin, M. R., Balderston, N. L. \u0026amp; Helmstetter, F. J. Prefrontal cortical regulation of fear learning. \u003cem\u003eTrends Neurosci.\u003c/em\u003e 37, 455\u0026ndash;464 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMilad, M. R. \u003cem\u003eet al.\u003c/em\u003e Recall of fear extinction in humans activates the ventromedial prefrontal cortex and hippocampus in concert. \u003cem\u003eBiol. Psychiatry\u003c/em\u003e 62, 446\u0026ndash;454 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoustafa, A. A. \u003cem\u003eet al.\u003c/em\u003e A model of amygdala\u0026ndash;hippocampal\u0026ndash;prefrontal interaction in fear conditioning and extinction in animals. \u003cem\u003eBrain Cogn.\u003c/em\u003e 81, 29\u0026ndash;43 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePic\u0026oacute;-P\u0026eacute;rez, M. \u003cem\u003eet al.\u003c/em\u003e Common and distinct neural correlates of fear extinction and cognitive reappraisal: A meta-analysis of fMRI studies. \u003cem\u003eNeurosci. Biobehav. Rev.\u003c/em\u003e 104, 102\u0026ndash;115 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRessler, K. J. Translating Across Circuits and Genetics Toward Progress in Fear- and Anxiety-Related Disorders. \u003cem\u003eAm. J. Psychiatry\u003c/em\u003e 177, 214\u0026ndash;222 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChikazoe, J., Lee, D. H., Kriegeskorte, N. \u0026amp; Anderson, A. K. Population coding of affect across stimuli, modalities and individuals. \u003cem\u003eNat. Neurosci.\u003c/em\u003e 17, 1114\u0026ndash;1122 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLindquist, K. A., Satpute, A. B., Wager, T. D., Weber, J. \u0026amp; Barrett, L. F. The Brain Basis of Positive and Negative Affect: Evidence from a Meta-Analysis of the Human Neuroimaging Literature. \u003cem\u003eCereb. Cortex N. Y. N 1991\u003c/em\u003e 26, 1910\u0026ndash;1922 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWen, Z. \u003cem\u003eet al.\u003c/em\u003e Distributed neural representations of conditioned threat in the human brain. \u003cem\u003eNat. Commun.\u003c/em\u003e 15, 2231 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMahmoudi, A., Takerkart, S., Regragui, F., Boussaoud, D. \u0026amp; Brovelli, A. Multivoxel Pattern Analysis for fMRI Data: A Review. \u003cem\u003eComput. Math. Methods Med.\u003c/em\u003e 2012, 961257 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeaverdyck, M. E., Lieberman, M. D. \u0026amp; Parkinson, C. Tools of the Trade Multivoxel pattern analysis in fMRI: a practical introduction for social and affective neuroscientists. \u003cem\u003eSoc. Cogn. Affect. Neurosci.\u003c/em\u003e 15, 487\u0026ndash;509 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKragel, P. A. \u0026amp; LaBar, K. S. Advancing emotion theory with multivariate pattern classification. \u003cem\u003eEmot. Rev. J. Int. Soc. Res. Emot.\u003c/em\u003e 6, 160\u0026ndash;174 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou, F. \u003cem\u003eet al.\u003c/em\u003e A distributed fMRI-based signature for the subjective experience of fear. \u003cem\u003eNat. Commun.\u003c/em\u003e 12, 6643 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStelzer, J., Chen, Y. \u0026amp; Turner, R. Statistical inference and multiple testing correction in classification-based multi-voxel pattern analysis (MVPA): Random permutations and cluster size control. \u003cem\u003eNeuroImage\u003c/em\u003e 65, 69\u0026ndash;82 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBuckner, R. L., Krienen, F. M., Castellanos, A., Diaz, J. C. \u0026amp; Yeo., B. T. The organization of the human cerebellum estimated by intrinsic functional connectivity. \u003cem\u003eJ. Neurophysiol.\u003c/em\u003e 106, 2322\u0026ndash;2345 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYeo, B. T. T. \u003cem\u003eet al.\u003c/em\u003e The organization of the human cerebral cortex estimated by intrinsic functional connectivity. \u003cem\u003eJ. Neurophysiol.\u003c/em\u003e 106, 1125\u0026ndash;1165 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavignon, L.-M. \u003cem\u003eet al.\u003c/em\u003e Associations between past and current use of oral contraceptives and fear regulation. \u003cem\u003eBiol. Psychiatry Cogn. Neurosci. Neuroimaging\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.bpsc.2025.09.018\u003c/span\u003e\u003cspan address=\"10.1016/j.bpsc.2025.09.018\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2025) doi:10.1016/j.bpsc.2025.09.018 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMonti, M. M. Statistical Analysis of fMRI Time-Series: A Critical Review of the GLM Approach. \u003cem\u003eFront. Hum. Neurosci.\u003c/em\u003e 5, (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGenovese, C. R., Lazar, N. A. \u0026amp; Nichols, T. Thresholding of statistical maps in functional neuroimaging using the false discovery rate. \u003cem\u003eNeuroImage\u003c/em\u003e 15, 870\u0026ndash;878 (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoeller, J. R. \u0026amp; Habeck, C. G. Reciprocal Benefits of Mass-Univariate and Multivariate Modeling in Brain Mapping: Applications to Event-Related Functional MRI, H215O-, and FDG-PET. \u003cem\u003eInt. J. Biomed. Imaging\u003c/em\u003e 2006, 079862 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLindquist, K., Satpute, A., Wager, T., Weber, J. \u0026amp; Barrett, L. Neurosynth: 25631056. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://neurosynth.org/studies/25631056/\u003c/span\u003e\u003cspan address=\"https://neurosynth.org/studies/25631056/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFan, L. \u003cem\u003eet al.\u003c/em\u003e The Human Brainnetome Atlas: A New Brain Atlas Based on Connectional Architecture. \u003cem\u003eCereb. Cortex N. Y. N 1991\u003c/em\u003e 26, 3508\u0026ndash;3526 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBobbitt, Z. Fisher Z-Transformation: Definition \u0026amp; Example. \u003cem\u003eStatology\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.statology.org/fisher-z-transformation/\u003c/span\u003e\u003cspan address=\"https://www.statology.org/fisher-z-transformation/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXie, T. \u003cem\u003eet al.\u003c/em\u003e The case for hemispheric lateralization of the human amygdala in fear processing. \u003cem\u003eMol. Psychiatry\u003c/em\u003e 30, 2252\u0026ndash;2259 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGogolla, N. The insular cortex. \u003cem\u003eCurr. Biol.\u003c/em\u003e 27, R580\u0026ndash;R586 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStevens, F., Hurley, R. \u0026amp; Taber, K. Anterior Cingulate Cortex: Unique Role in Cognition and Emotion. \u003cem\u003eJ. Neuropsychiatry Clin. Neurosci.\u003c/em\u003e 23, (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeiner, K. S. \u0026amp; Zilles, K. The anatomical and functional specialization of the fusiform gyrus. \u003cem\u003eNeuropsychologia\u003c/em\u003e 83, 48\u0026ndash;62 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMion, M. \u003cem\u003eet al.\u003c/em\u003e What the left and right anterior fusiform gyri tell us about semantic memory. \u003cem\u003eBrain J. Neurol.\u003c/em\u003e 133, 3256\u0026ndash;3268 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFleury, M. N. \u003cem\u003eet al.\u003c/em\u003e Long-term memory plasticity in a decade-long connectivity study post anterior temporal lobe resection. \u003cem\u003eNat. Commun.\u003c/em\u003e 16, 692 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchwab, S. \u003cem\u003eet al.\u003c/em\u003e Functional Connectivity Alterations of the Temporal Lobe and Hippocampus in Semantic Dementia and Alzheimer\u0026rsquo;s Disease. \u003cem\u003eJ. Alzheimers Dis.\u003c/em\u003e 76, 1461\u0026ndash;1475 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaschereau-Dumouchel, V. \u003cem\u003eet al.\u003c/em\u003e Interaction between the prefrontal and visual cortices supports subjective fear. \u003cem\u003ePhilos. Trans. R. Soc. B Biol. Sci.\u003c/em\u003e 379, 20230245 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBilalić, M., Langner, R., Ulrich, R. \u0026amp; Grodd, W. Many faces of expertise: fusiform face area in chess experts and novices. \u003cem\u003eJ. Neurosci. Off. J. Soc. Neurosci.\u003c/em\u003e 31, 10206\u0026ndash;10214 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTarr, M. J. \u0026amp; Gauthier, I. FFA: a flexible fusiform area for subordinate-level visual processing automatized by expertise. \u003cem\u003eNat. Neurosci.\u003c/em\u003e 3, 764\u0026ndash;769 (2000).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcGugin, R. W., Gatenby, J. C., Gore, J. C. \u0026amp; Gauthier, I. High-resolution imaging of expertise reveals reliable object selectivity in the fusiform face area related to perceptual performance. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 109, 17063\u0026ndash;17068 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcGugin, R. W., Van Gulick, A. E. \u0026amp; Gauthier, I. Cortical Thickness in Fusiform Face Area Predicts Face and Object Recognition Performance. \u003cem\u003eJ. Cogn. Neurosci.\u003c/em\u003e 28, 282\u0026ndash;294 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJung, Y.-H. \u003cem\u003eet al.\u003c/em\u003e Altered Amygdala Resting-State Functional Connectivity and Hemispheric Asymmetry in Patients With Social Anxiety Disorder. \u003cem\u003eFront. Psychiatry\u003c/em\u003e 9, 164 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCanchola, E., Matsumura, P. D. \u0026amp; Rivera, R. M. Fusiform Gyrus: The Sorcerer of the Brain and Architect of Subjective Reality. \u003cem\u003eMed. - Clin. - Res.\u003c/em\u003e 1, 49\u0026ndash;51 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKryklywy, J. H., Forys, B. J., Vieira, J. B., Quinlan, D. J. \u0026amp; Mitchell, D. G. V. Dissociating representations of affect and motion in visual cortices. \u003cem\u003eCogn. Affect. Behav. Neurosci.\u003c/em\u003e 23, 1322\u0026ndash;1345 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSack, A. T., Kohler, A., Linden, D. E. J., Goebel, R. \u0026amp; Muckli, L. The temporal characteristics of motion processing in hMT/V5+: Combining fMRI and neuronavigated TMS. \u003cem\u003eNeuroImage\u003c/em\u003e 29, 1326\u0026ndash;1335 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFurl, N., Henson, R. N., Friston, K. J. \u0026amp; Calder, A. J. Top-Down Control of Visual Responses to Fear by the Amygdala. \u003cem\u003eJ. Neurosci.\u003c/em\u003e 33, 17435\u0026ndash;17443 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePalejwala, A. H. \u003cem\u003eet al.\u003c/em\u003e Anatomy and white matter connections of the fusiform gyrus. \u003cem\u003eSci. Rep.\u003c/em\u003e 10, 13489 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLithari, C., Moratti, S. \u0026amp; Weisz, N. Limbic areas are functionally decoupled and visual cortex takes a more central role during fear conditioning in humans. \u003cem\u003eSci. Rep.\u003c/em\u003e 6, 29220 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, W. \u0026amp; Keil, A. Sensing fear: fast and precise threat evaluation in human sensory cortex. \u003cem\u003eTrends Cogn. Sci.\u003c/em\u003e 27, 341\u0026ndash;352 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBayle, D. J., Henaff, M.-A. \u0026amp; Krolak-Salmon, P. Unconsciously Perceived Fear in Peripheral Vision Alerts the Limbic System: A MEG Study. \u003cem\u003ePLOS ONE\u003c/em\u003e 4, e8207 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavis, T. \u003cem\u003eet al.\u003c/em\u003e What do differences between multi-voxel and univariate analysis mean? How subject-, voxel-, and trial-level variance impact fMRI analysis. \u003cem\u003eNeuroImage\u003c/em\u003e 97, 271\u0026ndash;283 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHebart, M. N. \u0026amp; Baker, C. I. Deconstructing multivariate decoding for the study of brain function. \u003cem\u003eNeuroImage\u003c/em\u003e 180, 4\u0026ndash;18 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu, J. \u003cem\u003eet al.\u003c/em\u003e Cross-cohort replicability and generalizability of connectivity-based psychometric prediction patterns. \u003cem\u003eNeuroImage\u003c/em\u003e 262, 119569 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYeretzian, S. T., Sahakyan, Y., Kozloff, N. \u0026amp; Abrahamyan, L. Sex differences in the prevalence and factors associated with anxiety disorders in Canada: A population-based study. \u003cem\u003eJ. Psychiatr. Res.\u003c/em\u003e 164, 125\u0026ndash;132 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBauer, E. P. Sex differences in fear responses: Neural circuits. \u003cem\u003eNeuropharmacology\u003c/em\u003e 222, 109298 (2023).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"translational-psychiatry","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"tp","sideBox":"Learn more about [Translational Psychiatry](http://www.nature.com/tp/)","snPcode":"41398","submissionUrl":"https://mts-tp.nature.com/cgi-bin/main.plex","title":"Translational Psychiatry","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Subjective fear, fMRI, Mass-univariate analyses, MVPA, canonical system","lastPublishedDoi":"10.21203/rs.3.rs-9456965/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9456965/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe neural basis of the subjective experience of fear remains incompletely understood. Although fear has traditionally been associated with a circumscribed set of brain regions, recent findings have challenged this view by suggesting that fear-related representations may be widely distributed across the entire brain. In the present study, we investigate the validity of this claim by testing whether a common neural network can be reliably identified across three independent functional MRI datasets (total \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;251) probing distinct fear domains: (1) fear of personally relevant animals, (2) fear of diverse threatening images and (3) fear of conditioned threat cues. Using a combination of mass-univariate analyses and multivariate pattern analysis, we identified a core network consistently engaged across all three datasets, encompassing the anterior cingulate gyrus, the insular cortex, and lateral occipitotemporal regions. Notably, while fear could be predicted from a distributed pattern within each dataset, a shared network could be identified across datasets and analytical approaches. These findings suggest that fear representations are neither strictly localized nor fully diffuse, but instead rely on a reproducible set of regions embedded within broader brain-wide patterns. We discuss how methodological decisions influence conclusions about the spatial organization of fear in the brain, and consider the implications of these results for translational and clinical efforts aimed at modulating pathological fear.\u003c/p\u003e","manuscriptTitle":"Is the representation of fear distributed across the whole brain?","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-11 08:23:18","doi":"10.21203/rs.3.rs-9456965/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-04-30T11:14:51+00:00","index":2,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-04-30T06:36:41+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewersInvited","content":"","date":"2026-04-29T19:34:20+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-21T11:34:31+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-21T06:44:15+00:00","index":"","fulltext":""},{"type":"submitted","content":"Translational Psychiatry","date":"2026-04-20T19:16:30+00:00","index":"","fulltext":""},{"type":"checksFailed","content":"","date":"2026-04-20T16:14:35+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"translational-psychiatry","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"tp","sideBox":"Learn more about [Translational Psychiatry](http://www.nature.com/tp/)","snPcode":"41398","submissionUrl":"https://mts-tp.nature.com/cgi-bin/main.plex","title":"Translational Psychiatry","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"34de46c6-8cce-485d-9258-658737457355","owner":[],"postedDate":"May 11th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-04-30T11:14:51+00:00","index":2,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-04-30T06:36:41+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewersInvited","content":"6","date":"2026-04-29T19:34:20+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":67269073,"name":"Health sciences/Diseases/Psychiatric disorders"},{"id":67269074,"name":"Biological sciences/Neuroscience"}],"tags":[],"updatedAt":"2026-05-11T08:23:19+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-11 08:23:18","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9456965","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9456965","identity":"rs-9456965","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}