henhaobio
Sign in

bioRxivNeurosciencePreprint

Dissociable Neural Responses to Conditioned Social Threats are Modulated by Spatial Proximity in PTSD and MDD

N. C. SteeleiD, Leonel Rangel-Jimenez, Dash A. Watts, LeeAnne TunstalliD, Jenna BeakasiD, Ahmed Hussain, Ashley Huggins, Delin SuniD, Kevin S. LaBariD, Rajendra A. MoreyiD

PDFDOI
In-depth full-text digest available · Jump to digest ↓

Abstract

The brains response to potential threats is shaped by the egocentric spatiotemporal distance to the threat. While distal threats recruit evaluative, cognitive-fear networks for strategic planning, proximal threats engage evolutionarily conserved reactive-fear circuitry to mount immediate defensive responses. Understanding this dual neural architecture is clinically relevant, as the spatial proximity to a traumatic event predicts subsequent psychiatric symptom severity and recovery. Despite the well-characterized disruptions to threat circuitry in posttraumatic stress disorder (PTSD) and major depressive disorder (MDD), how these networks respond to threats at distinct spatial distances has yet to be investigated in these disorders. Utilizing 3D virtual reality technology, we implemented a spatially-modulated fear conditioning paradigm by presenting human avatars at proximal (peripersonal) and distal (extrapersonal) distances, paired with shock, to trauma-exposed participants (n = 50) during functional MRI (fMRI). We modeled associations between PTSD and MDD symptom severity and task-evoked hemodynamic responses within cognitive-fear and reactive-fear threat networks. Our results support a transdiagnostic disruption of thalamic responses to proximal threats, driven by heightened activation to the safety stimulus and stronger deactivation to the threat stimulus. MDD showed a disorder-specific effect of disrupted activity across cognitive-fear and social cognition regions in response to proximal social threats, and illness severity-by-proximity effects across motor regions. PTSD symptom severity was uniquely associated with hyperactivity of the amygdala to distal threats. Transdiagnostic disruption of connectivity between the dorsal precuneus and several subcortical structures followed a disorder-specific gradient across the anterior-posterior axis. While thalamic disruptions to threats represent a shared transdiagnostic effect, PTSD and MDD are distinguished by differences in amygdala hyperactivation and cognitive-fear network hypoactivation, respectively.

In-depth summary

Claude Opus 5.5 · xhigh effort · 7 Nov 2026 · from the full text of version 1 (16 Jul 2026) · CC BY 4.0

In 50 trauma-exposed veterans, thalamic hypoactivity to near virtual-reality threats tracked both PTSD and MDD severity, while amygdala hyperactivity to far threats tracked only PTSD severity.

How close someone is to a trauma predicts their later symptom severity, yet no one had tested how threat networks tuned to distal (cognitive-fear) and proximal (reactive-fear) threats respond in PTSD and MDD. The study takes a 3D virtual-reality paradigm with human avatars, which Faul et al. earlier ran in a healthy sample, into a trauma-exposed cohort scanned with fMRI; most earlier spatial threat studies used chase games built from geometric shapes.

Key results

  1. SuggestedIn both disorders, higher symptom severity went with a weaker thalamic threat response (CS+ > CS-) to near avatars in late acquisition. The effect localized to the ventral posterolateral (VPL) nucleus, and for PTSD it extended into the anterior pulvinar.

    PTSD (CAPS-5) and MDD (BDI-II) severity were modeled separately, with age and sex fixed and scanner and display as random intercepts. PTSD symptom clusters explained 43.2% of thalamic variance; avoidance took 39.6% of R^2 and dominated every submodel. The abstract attributes the shared effect to more activation to the safety stimulus and deeper deactivation to the threat stimulus. · Abstract; PTSD Severity Exhibits Distinct Associations…; MDD Severity Exhibits Distinct Associations with…; Thalamic Hypoactivity is a Transdiagnostic…

  2. SuggestedOnly PTSD severity was linked to amygdala hyperactivity. Higher CAPS-5 scores predicted a larger CS+Far > CS-Far amygdala response in late acquisition, extending into the anterior hippocampus, and the authors describe the effect as selective for distal threats.

    In a dominance analysis of the amygdala cluster (R^2 = 0.47), hyperarousal (32.2%) and negative mood/cognition (31.5%) contributed most, ahead of re-experiencing (18.7%) and avoidance (17.7%). No amygdala activation effect is reported for MDD severity. · PTSD Severity Exhibits Distinct Associations…; Amygdala Hyperreactivity is Specific to PTSD; Conclusion; Figure legends

  3. SuggestedMDD severity tracked reduced near-threat responses across cognitive-fear and reactive-fear regions (midbrain early; VPL thalamus, dorsolateral prefrontal cortex (dlPFC), posterior cingulate cortex (PCC) and angular gyrus late) and premotor/motor cortex for far threats, via higher CS- and lower CS+ activity.

    In late acquisition, severity-by-threat interactions appeared in right dlPFC, right posterior supramarginal gyrus, anterior-dorsal precuneus and medial pulvinar. Severity-by-proximity effects appeared in right cerebellum VIII (near > far, early) and primary motor cortex (far > near, late). PTSD severity instead showed early dlPFC hyperactivity to far versus near stimuli. · MDD Severity Exhibits Distinct Associations with…; Transdiagnostic and Disorder-Specific Activity…; Dorsolateral Prefrontal Cortex

  4. SuggestedIn early acquisition, connectivity between the dorsal precuneus and subcortical seeds followed an anterior-posterior gradient. The PTSD effect stayed in the posterior-dorsal precuneus, while MDD effects spanned the whole anterior-posterior axis and changed direction with seed and distance.

    gPPI seed-to-voxel models. With PTSD severity: stronger posterior hippocampus coupling for far threats (896 mm^3, max-Z = 4.45, pbonferroni < 0.05; hyperarousal 50% of R^2 = 0.34). With MDD severity: stronger amygdala coupling for far threats (600 mm^3, max-Z = 4.33) and weaker periaqueductal gray (PAG) coupling for near threats (680 mm^3, max-Z = -6.26). · PTSD Severity Exhibits Distinct Associations…; MDD Severity Exhibits Distinct Associations with…; Dorsal Precuneus

  5. ShownParticipants learned which avatars predicted shock at both distances, and they judged near shock-paired avatars more dangerous than far ones. Symptom severity showed no robust link to behavioral threat discrimination.

    Shock expectancy (n = 34): CS+ > CS- (F = 13.34, p < 0.001), with no interaction with distance (F = 2.824, p = 0.102). Danger ratings (n = 50): threat-by-proximity F(1, 46) = 5.21, p = 0.027; CS+Near M = 7.43 vs CS+Far M = 6.13 (t = 4.71, p < 0.001); CS-Near vs CS-Far t = 0.48, p = 0.63. A nominal PTSD-arousal correlation (r = 0.35) did not survive covariate adjustment (p = 0.18). · US Expectancy Ratings; Shock Expectancy and Post-Scan Stimulus Ratings…; Figure legends

Methods, data and reproducibility

Fifty trauma-exposed veterans underwent fMRI during 3D virtual-reality fear conditioning with four male avatars at 1.0 m or 5.0 m; two of them were paired with wrist shock on 60% of trials. AFNI 3dLMEr mixed models related CAPS-5 (PTSD) and BDI-II (MDD) severity to activation and to generalized psychophysiological interaction (gPPI) connectivity in early and late acquisition.

  • Trauma-exposed post-2001 military veterans recruited through the Durham VA Health Care System; fMRI during virtual-reality fear acquisition, with CAPS-5 and BDI-II symptom severity · human · n = 50
  • Subset with in-scanner shock expectancy ratings (ratings missing for 16 participants after button-box failures) · human · n = 34
Methods & tools
AFNI 3dLMEr, gPPI, aCompCor, AAL atlas, Harvard-Oxford probabilistic atlas, Morel atlas, BIOPAC MP-150 with STM200 stimulator, MagnetVision 3D television, Nordic Neuro VisualSystem HD goggles, CAPS-5, BDI-II, dominance analysis
Reproducibility gaps
The text has no code or data availability statement. It does not describe fMRI acquisition, preprocessing or activation thresholds, and it cites Tables 1 to 3 and the supplementary material without including their contents.
Limitations (5)
  • The cohort was military veterans with mostly combat-related index traumas, so it is unknown whether the results hold for civilians or other trauma types (such as interpersonal violence or chronic developmental trauma). · stated by the authors
  • The MDD severity effects may apply only to trauma-exposed people: trauma is common in MDD (50-80%), but a significant subgroup of patients has none. · stated by the authors
  • Psychotropic medication is a potential confound, which the authors argue is mitigated by capping current users at 25% of the sample. · stated by the authors
  • The Results text reports stronger posterior hippocampus-precuneus connectivity with PTSD severity, but the Figure 3 legend says weaker. The Figure 4 legend names the posterior hippocampus where the text reports amygdala-precuneus coupling for MDD. · noted in this summary
  • This is a single cross-sectional cohort of 50 with no replication. PTSD and MDD severity were modeled separately (the joint model appears only in Table S3), so disorder specificity is not tested head to head, and the text does not describe voxelwise thresholds for the activation results. · noted in this summary

Read this if you study threat imminence or fear learning in PTSD or depression and want a virtual-reality fMRI paradigm that separates near from far social threats.

Citations
1

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISSun '20Tan '19Chiba '20Hauner '24Dopfel '19Jia '20
74% semantic matchNeuroscience
Delin Sun, Andrea L. Gold, Chelsea A. Swanson+21 more·Translational Psychiatry·Feb 10, 2020· 23 citations