Huberman LabHuberman Lab

Essentials: Compulsive Behaviors & Deep Brain Stimulation | Dr. Casey Halpern

Andrew Huberman and Dr. Casey Halpern on deep brain stimulation reveals circuits driving compulsions, cravings, and treatment innovation.

Andrew HubermanhostDr. Casey Halpernguest
May 7, 202633mWatch on YouTube ↗
What neurosurgeons do vs. psychiatry/neurologyDeep brain stimulation as electrical “medication”OCD definition, spectrum, and treatment ladderPrefrontal/orbitofrontal cortex and basal ganglia circuitsVentral striatum/nucleus accumbens and urges despite riskBinge eating disorder, addiction parallels, and craving signalsTMS and MRI-guided focused ultrasound (ablation vs modulation)Invasive monitoring (stereo-EEG concepts) for mental healthAwareness limits in severe/refractory compulsive disordersMachine learning/wearables for predicting high-risk states
AI-generated summary based on the episode transcript.

In this episode of Huberman Lab, featuring Andrew Huberman and Dr. Casey Halpern, Essentials: Compulsive Behaviors & Deep Brain Stimulation | Dr. Casey Halpern explores deep brain stimulation reveals circuits driving compulsions, cravings, and treatment innovation Neurosurgery for psychiatric and behavioral disorders focuses on modulating specific brain circuits using deep brain stimulation (DBS) or small targeted ablations when medications and therapy fail.

At a glance

WHAT IT’S REALLY ABOUT

Deep brain stimulation reveals circuits driving compulsions, cravings, and treatment innovation

  1. Neurosurgery for psychiatric and behavioral disorders focuses on modulating specific brain circuits using deep brain stimulation (DBS) or small targeted ablations when medications and therapy fail.
  2. OCD is framed as a spectrum where adaptive traits become disabling when obsessions and compulsions are uncontrollable, with first-line treatments including SSRIs/tricyclics and exposure-response prevention therapy.
  3. Key implicated circuitry includes hyperactivity in prefrontal/orbitofrontal cortex and downstream basal ganglia/ventral striatum pathways, especially reward/compulsion gating regions that can drive urges despite risk.
  4. Halpern’s research aims to identify symptom-specific neural signatures (e.g., “obsession” or “craving” signals) to improve electrode placement and enable closed-loop stimulation that responds to pathological states.
  5. Non-invasive approaches like TMS and MRI-guided focused ultrasound are promising and increasingly FDA-cleared for specific indications, but are limited by incomplete mechanistic understanding and uncertain targets for many psychiatric conditions.

IDEAS WORTH REMEMBERING

5 ideas

DBS is best understood as targeted circuit modulation, not “brain surgery that fixes behavior.”

Halpern describes DBS as implanting a delivery tool where the therapy is precisely delivered electrical stimulation; tiny shifts in where/what you stimulate can cause immediate emotional or behavioral effects that can be turned off.

OCD differs from everyday obsessiveness by loss of control and functional impairment.

He notes that obsession/compulsiveness can be advantageous in surgeons or CEOs, but becomes OCD when it’s uncontrollable, persistent, and disruptive—especially in the severe, treatment-refractory cases that reach surgical clinics.

Current severe-OCD surgical options work in only about half of patients—and responders often remain symptomatic.

Halpern cites roughly ~50% responder rates for invasive interventions like DBS or capsulotomy, motivating efforts to find more symptom-specific targets and better predictive markers of who will benefit.

Compulsion and addiction-like behavior share a common computation: ‘urge despite risk.’

He links OCD checking/cleaning rituals, drug seeking, and binge/purge behaviors to shared circuitry that can override judgment, highlighting the ventral striatum/nucleus accumbens as a key node for gating reward pursuit.

A major research frontier is finding ‘symptom signatures’ (obsession/craving signals) in humans to guide placement and closed-loop therapy.

Using intraoperative recordings (single-unit) and implanted devices (population signals), his group attempts to detect neural activity that precedes a binge or obsession so stimulation can be triggered only when pathological states arise.

WORDS WORTH SAVING

5 quotes

Deep brain stimulation is a procedure where we have to place a, uh, a very thin wire that's insulated deep into, uh, a part of the brain that's involved in Parkinson's disease, for example. Uh, but that's actually not the therapy. The therapy is delivering electrical stimulation through the tip of that wire...

Dr. Casey Halpern

I consider OCD to be a, a spectrum disorder in a way.

Dr. Casey Halpern

The issue is if you have an urge for re- a reward that either puts you or somebody else at risk, it's probably a reward we shouldn't have.

Dr. Casey Halpern

Craving. So craving is a term that, you know, uh, there's probably other terms we could use, by the way, but that, that's the term we've chosen to use for a number of reasons. One, because people relate with that term.

Dr. Casey Halpern

I've always said we have to get in the brain before we get out of it, and if we get in the brain and understand what these signals look like, we'll know what those non-invasive signals are.

Dr. Casey Halpern

QUESTIONS ANSWERED IN THIS EPISODE

5 questions

When you say OCD is a ‘spectrum,’ what clinical features mark the line between adaptive compulsiveness and diagnosable disorder in your surgical candidates?

Neurosurgery for psychiatric and behavioral disorders focuses on modulating specific brain circuits using deep brain stimulation (DBS) or small targeted ablations when medications and therapy fail.

You mention ~50% responder rates for severe OCD surgery—what distinguishes responders from non-responders (symptom subtype, circuitry, comorbidities, imaging markers)?

OCD is framed as a spectrum where adaptive traits become disabling when obsessions and compulsions are uncontrollable, with first-line treatments including SSRIs/tricyclics and exposure-response prevention therapy.

For nucleus accumbens/ventral striatum targeting, what specific behavior or neural pattern tells you you’re in the right spot during an awake procedure?

Key implicated circuitry includes hyperactivity in prefrontal/orbitofrontal cortex and downstream basal ganglia/ventral striatum pathways, especially reward/compulsion gating regions that can drive urges despite risk.

In binge eating experiments, what is the earliest measurable signal (brain, eye-tracking, physiology) that reliably precedes ‘the bite,’ and how much lead time does it provide for intervention?

Halpern’s research aims to identify symptom-specific neural signatures (e.g., “obsession” or “craving” signals) to improve electrode placement and enable closed-loop stimulation that responds to pathological states.

TMS is FDA-approved for OCD and nicotine addiction—what do you see as the biggest limitation today: coil targeting, dosing schedules, patient selection, or lack of circuit biomarkers?

Non-invasive approaches like TMS and MRI-guided focused ultrasound are promising and increasingly FDA-cleared for specific indications, but are limited by incomplete mechanistic understanding and uncertain targets for many psychiatric conditions.

Chapter Breakdown

Meet Dr. Casey Halpern: functional neurosurgery and why DBS is uniquely powerful

Huberman introduces Dr. Casey Halpern and frames neurosurgeons as clinicians who can directly interact with brain circuits to treat disease. Halpern sets up how functional neurosurgery—especially deep brain stimulation (DBS)—can both relieve symptoms and reveal how emotion and behavior circuits work in humans.

What neurosurgeons do—and what makes deep brain stimulation different from other brain procedures

Halpern distinguishes the broad scope of neurosurgery (tumors, aneurysms, trauma, spine, peripheral nerves) from subspecialized functional neurosurgery. He explains DBS as implanting a thin wire with multiple contacts to deliver electricity like a targeted, adjustable “medication.”

Side effects that teach: emotion circuits revealed during stimulation mapping

Halpern describes how stimulating near a target can cause brief effects like laughter or panic—temporary phenomena that can be shut off immediately. These observations helped broaden DBS beyond movement disorders to psychiatric symptoms and comorbidities, suggesting overlap between motor and limbic circuits.

Defining OCD: spectrum vs disorder, and how clinicians decide when it’s pathological

Halpern frames OCD as existing on a spectrum—traits like obsessionality/compulsivity can be adaptive in certain professions, but become disabling when uncontrollable. He emphasizes that the surgical population is typically the most severe and treatment-refractory group, motivating circuit-level interventions.

Standard OCD treatments and why some patients remain refractory

The discussion reviews first-line OCD treatments—primarily serotonin-targeting medications and exposure-response prevention therapy. Halpern notes that despite strong options, a substantial minority do not achieve adequate relief, leading to consideration of more invasive approaches.

Surgical options for severe OCD: DBS vs capsulotomy (ablation) and the response ceiling

Halpern contrasts DBS with ablative approaches like capsulotomy, noting that small lesions (3–4 mm) can help with surprisingly few obvious deficits, but are irreversible. He also highlights a major limitation: even with surgery, response rates and degree of symptom relief are often modest, pushing the field toward more precise targeting.

OCD circuitry: hyperactive frontal cortex and basal ganglia loops linked to compulsion

Halpern outlines OCD as involving both cortical and subcortical dysfunction—particularly hyperactivity in prefrontal/orbitofrontal regions and their loops through the basal ganglia. He points to ventral striatum circuitry as central to compulsive behavior, connecting OCD patterns to addiction and eating disorders via “urge despite risk.”

Nucleus accumbens: reward, habit, and the biology of ‘craving’

The nucleus accumbens is presented as a hub within reward circuitry whose function can be “hijacked” by repeated exposure to powerful rewards (drugs, binge foods). Halpern clarifies the therapeutic goal is not to eliminate normal desire, but to disrupt pathological, risky, repetitive urges.

From tremor cells to craving cells: using intraoperative recordings to find symptom-specific signals

Halpern explains how DBS targeting in Parkinson’s uses microelectrode recordings—converting neural firing into sound to identify tremor-related activity. He describes translating that logic to psychiatry and eating behavior by seeking “craving” or “obsession” signals during awake procedures to guide precise electrode placement.

Non-invasive neuromodulation: TMS today and focused ultrasound as a next frontier

Huberman and Halpern evaluate non-invasive options, acknowledging both promise and mechanistic uncertainty. Halpern notes TMS is FDA-approved for depression and also for OCD and nicotine addiction, and argues it can help define circuits that might later justify invasive therapy or guide better targets.

MRI-guided focused ultrasound: lesioning, modulation, and the target-discovery problem

Halpern describes MRI-guided focused ultrasound as an FDA-approved, incisionless method to ablate tissue for tremor, producing dramatic benefit. Extending it to psychiatric disease is limited less by technology than by uncertainty about where to intervene; the field needs better circuit mapping to identify new, effective targets beyond classic capsulotomy zones.

Why awareness helps some—but fails in the most severe cases: provoking binges and decoding signals

Huberman proposes that improved self-awareness before cravings might prevent episodes; Halpern agrees but stresses that surgical candidates are often already highly aware and still lose control. He describes lab paradigms that provoke binge-related mood states while recording brain signals, using synchronized video and eye tracking to pinpoint neural changes immediately before eating.

AI, wearables, and scalable prediction of impulsive/compulsive episodes

The conversation turns to whether machine learning could predict high-risk states (e.g., suicidality, impulsive behavior) before conscious awareness, using signals like voice, sleep, and physiology. Halpern argues scalable solutions are essential given the population-level burden, but they must be grounded in real neural signal understanding to avoid ineffective “gadget” interventions.

Closing: the future of circuit-based therapies and inspiring the next generation

Huberman thanks Halpern and emphasizes the importance of the work at the cutting edge of brain repair and understanding. The episode ends with encouragement for future physicians/neurosurgeons and recognition of the broader mission: translating precise circuit interventions into better, more accessible therapies.

EVERY SPOKEN WORD

Install uListen for AI-powered chat & search across the full episode — Get Full Transcript

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

Add to Chrome