In a rare clinical case involving a man with drug-resistant epilepsy and severe post-traumatic stress disorder (PTSD), selectively inactivating part of his right amygdala reduced his tendency to perceive ambiguous faces as fearful and was followed by a sharp drop in trauma symptoms. Published in Nature Communications, the study tracked brain signals and emotion perception before and immediately after the awake bedside procedure. Although drawn from a single patient, the findings offer unusual causal evidence that right-amygdala activity helps sustain threat-focused biases and PTSD symptoms.

Threat bias and the amygdala in PTSD

PTSD is a psychiatric condition that can develop after severe trauma, marked by intrusive memories, avoidance of trauma reminders, negative shifts in mood and thinking, and persistent physical jumpiness. A central feature of the disorder is negative interpretation bias, the tendency to read neutral or ambiguous cues, such as an unreadable facial expression, as threatening. Seeing everyday surroundings as dangerous keeps fear responses active and makes emotions harder to regulate.

Research into this threat bias often focuses on the amygdala, a pair of almond-shaped neuron clusters deep inside the brain’s left and right temporal lobes, roughly behind the temples. The amygdala helps detect danger and coordinate fear responses, and brain scans consistently show heightened amygdala reactivity to fearful or neutral sights in people with PTSD. Standard brain imaging, however, only captures correlations, leaving uncertain whether abnormal amygdala activity drives negative interpretation bias or merely accompanies it.

Studying localized brain damage or targeted neurosurgery offers a rare way to observe what happens when amygdala function is directly disrupted. In people without PTSD, bilateral amygdala damage selectively dulls recognition of fearful faces while leaving happy-face recognition intact. Meanwhile, as the authors of a 2026 narrative review argued, patients treated for limbic epilepsy, where seizures arise in deep emotion and memory structures such as the amygdala and neighboring seahorse-shaped hippocampus, offer a clinical window for linking psychiatric symptoms to specific brain circuits.

Pinpointing how to intervene in these deep brain circuits has been difficult without direct internal brain-wave markers or established targets for therapies that adjust neural activity. “There is a lack of validated biomarkers of intracranial activity that localize and monitor abnormal cognitive activity in PTSD,” senior study author Jon T. Willie, chief of functional and epilepsy surgery at the University of Texas at Austin’s Dell Medical School, told PsyPost. Willie also pointed to a “lack of understanding of lateralized functions of right versus left amygdala in PTSD” and a “lack of robustly validated therapeutic targets for surgery or neuromodulation for PTSD.”

In a 2020 case series co-authored by Willie, laser ablation of both the right amygdala and adjacent hippocampus for epilepsy eased PTSD symptoms, leaving uncertain whether relief stemmed from disrupting the amygdala itself or from removing hippocampal tissue and stopping seizures. Led by first author Tao Xie alongside Willie and co-senior author Peter Brunner, the researchers studied a rare patient with epilepsy and chronic PTSD who underwent isolated right-amygdala ablation two months before hippocampal surgery, separating the amygdala’s role from seizure control. The ablation was performed primarily to control his seizures, with the PTSD outcomes tracked as a secondary question.

Mapping emotion and ablating the right amygdala

To perform that first-stage procedure, doctors used stereoelectroencephalography-guided radiofrequency ablation. Thin wire electrodes are implanted deep into the brain to record electrical activity and pinpoint seizure sources. High-frequency current then heats selected contacts to roughly 58 to 60 degrees Celsius, permanently inactivating a three- to seven-millimeter tissue zone while the patient stays awake.

The patient was a 41-year-old man with typical intellectual ability, drug-resistant temporal lobe epilepsy, and severe PTSD from trauma during adolescence. For baseline comparison, the team recruited 16 temporal lobe epilepsy patients undergoing deep-brain electrode monitoring at the same hospital, none of whom had PTSD. The group included seven men and nine women, with an average age of 37. De-identified behavioral and brain-recording data and analysis code for the patient were posted publicly on the Open Science Framework.

Seven electrode contacts were placed in the patient’s right amygdala and three in his left amygdala, alongside standard scalp electrodes. Early in hospitalization, antiseizure medications were briefly lowered to capture spontaneous right-hippocampus seizures, then fully restarted more than seven hours before testing began, with no seizures occurring during any task, stimulation, or ablation.

Before the ablation, the patient completed 252 trials of a facial emotion discrimination task, while the 16 controls completed roughly 260 trials on average. Participants viewed faces for one second morphed from 0 percent fearful (fully happy) to 100 percent fearful, with 50 percent representing an ambiguous expression, then classified each face as fearful or happy and rated their confidence. Electrodes simultaneously recorded late-phase brain waves 350 to 650 milliseconds after each face appeared, reflecting sustained emotional processing.

On the morning of the procedure, researchers mapped the patient’s amygdala by delivering one-second electrical pulses mixed randomly with sham trials across contact pairs in both hemispheres, asking him to report any physical or emotional sensations. That afternoon, doctors ablated tissue around five contacts in his right basal and lateral amygdala, and 20 minutes later he completed 336 trials of the face task.

Immediate shifts in fear perception and brain activity

Before ablation, the patient showed a marked negative interpretation bias compared with the 16 controls, judging ambiguous 50-percent faces as fearful notably more often. That difference reached statistical significance against the male controls but not the female controls. Group differences on happy faces were not statistically significant.

Scalp and right lateral amygdala recordings likewise showed a strong gap in late-phase electrical responses between fearful and happy faces, with most of his right lateral amygdala contacts exceeding the 90th percentile of controls; corresponding differences in his left amygdala and in control patients’ amygdalae were not statistically significant.

Pre-ablation electrical stimulation pointed to a similar left-right split. Mild currents of 1.5 to 2.0 milliamperes across three right-amygdala subregions reliably evoked tingling chills in his head, chest, and shoulder along with euphoria, whereas left-amygdala stimulation produced nothing up to 2.5 milliamperes and evoked a pleasant childhood memory at 3.0 milliamperes.

Post-procedure MRI scans showed that the lesion inactivated 47 percent of the right amygdala while sparing more than 99.9 percent of the right hippocampus, with electrode recordings confirming that the targeted tissue had stopped producing normal electrical activity. Retested 20 minutes later, the patient judged ambiguous and moderately fearful faces as fearful much less often across all four blocks, matching controls on intermediate expressions. Changes on moderately happy faces were not statistically significant, though compared with controls he rated unmistakably happy faces as fearful substantially more often and unmistakably fearful faces less often.

His decisions also slowed after ablation, with average reaction time rising from 0.7 to 1.1 seconds. He also grew less certain of his judgments overall, and his pre-ablation pattern of higher confidence when calling faces fearful than happy was no longer statistically significant.

Neurally, the late-phase electrical gap between fearful and happy faces vanished across the right amygdala and nearby right-hemisphere areas. Meanwhile, a new distinction emerged in his intact left amygdala alongside smaller responses to ambiguous faces. The authors suggest that silencing overactive threat signals from the right amygdala may have shifted emotion networks toward slower, more deliberate evaluation involving the left amygdala.

PTSD symptom relief independent of seizure control

To track clinical outcomes beyond the hospital stay, the patient completed validated questionnaires measuring PTSD, anxiety, and depression symptoms before hospitalization and two months after right-amygdala ablation. At that two-month mark, his PTSD score fell from 50 (severe) to 21, well below the diagnostic cutoff of 31 to 33, with gains across intrusive memories, avoidance, negative mood, and hyperarousal, alongside milder drops in anxiety and depression. His seizure rate stayed unchanged at roughly 10 per month during those two months, dropping below one per month only after a second-stage laser surgery on his right hippocampus and neighboring entorhinal cortex. That stability indicates his PTSD relief was not a byproduct of seizure control, though he noticed less heart racing and milder warning auras during seizures.

Notably, the patient was not undergoing structured talk therapy or taking PTSD-targeted psychiatric medications during the study period. “The remarkable result was evidence of immediate widespread changes in behavior, brain dynamics, and PTSD benefit without dependency on other traditional therapies such as medications or cognitive behavioral therapy,” Willie explained.

These findings align with an earlier brain-circuit and neurostimulation study of war veterans, which examined focal brain injuries and non-invasive magnetic brain stimulation in veterans rather than surgical ablation in a single patient, and similarly found that dampening amygdala-linked pathways eased PTSD symptoms. They also complement an earlier functional MRI study of identical twins, which likewise linked heightened amygdala reactivity to threat-related processing of ambiguous facial expressions in PTSD. That earlier study, however, measured correlational brain activation across twins rather than behavioral and symptom changes after surgical ablation in one individual.

“The novelty of the study is that this is the first clear demonstration of a specific causal role for the right amygdala in maintaining PTSD, indicating that therapies that reduce or block right amygdala could provide effective therapy for PTSD,” Willie said. At the same time, he noted that “this is an early feasibility study that could lead to future larger studies that target the right amygdala to treat PTSD.”

Caveats and next steps for research

There are a few things to keep in mind when interpreting these findings. The evidence comes from one patient whose brain circuits were shaped by years of drug-resistant epilepsy and antiseizure medications, so his neural organization may differ from that of most people with PTSD, and left-right amygdala specialization can vary across individuals.

“Readers should be cautioned that amygdala ablation has not been sufficiently studied to offer this widely as a therapy for PTSD,” Willie emphasized. “Better understanding of mechanisms and larger prospective controlled clinical trials are required to know the safety, feasibility, and effectiveness of amygdala interventions in different patient populations.”

Because awake neurosurgery cannot include a blinded sham procedure, expectations of relief, postoperative mood changes, or familiarity with repeating the face task 20 minutes after ablation may have influenced performance. In addition, the ablation slowed his judgments and shifted his ratings in opposite directions at both ends of the emotional continuum, suggesting that ablating amygdala tissue may blunt overall emotional calibration rather than removing threat bias alone.

Because the patient underwent hippocampal surgery two months later, the researchers could not isolate the durability of right-amygdala ablation beyond two months, though he still reported PTSD relief a year later. Future studies with larger samples, longer follow-up, and reversible techniques such as focused ultrasound or targeted electrical stimulation are needed to test whether quieting the right amygdala can safely treat severe PTSD.

To lay that groundwork, the team is investigating how the amygdala shapes the brain’s ability to recognize when a situation is safe rather than threatening. “Our next steps are to better define the mechanisms by which amygdala-focused interventions can be used to understand and treat PTSD,” Willie added. “We are studying neurosurgical epilepsy patients (in whom PTSD is also common) who undergo intracranial monitoring to understand basic mechanisms of fear and safety learning and how stimulation and ablation affect these processes and how this might be translated into therapies.”

The study, “Right amygdala ablation reduces maladaptive negative interpretation bias and symptoms in a patient with post-traumatic stress disorder,” was authored by Tao Xie, Sanne J. H. van Rooij, Sai Sun, Noah K. Bryson, Phillip Demarest, Haeorum Park, Luigi Maccotta, Shuo Wang, Peter Brunner, and Jon T. Willie.