[TMS thread] How to get fMRI data needed for Saint Protocol/ accelerated TMS/ tFUS for ADHD/excess slowwaves+ rumination in the brain?

#6

I make hardware and software for doing EEG neurofeedback. I’ve been working in this field for about 30 years. For many years I worked in a neurology clinic doing EEG bioeedback and quantitative EEG assessment and then I had my own practice
for several years. The effect size of brainwave biofeedback for ADHD is about equal to that of stimulant medication. However unlike the medication, the neurofeedback can make physical changes in the brain and electrical changes in the EEG patterns. With children who have a more plastic brain than adults, this kind of training is likely to produce permanent changes. Adults may need occasional booster sessions to keep the symptoms away.

If somebody is
charging you $4,500 a week to teach you how to change your brain function and you’re paying it, consider buying the Brooklyn Bridge. That’s way too much.

People typically need between 20 and 40 EEG Biofeedback sessions to fix ADHD. When I was practicing 15 years ago, $100 a session was a pretty typical cost.

The neurofeedback has wo parts. One is learning how to control your brainwaves. The other part is the effect of changing your brainwise several times a week in sessions for a long enough period time to rewire the circuitry associated with attention.

Think of it as other kinds of exercise. The first thing you do is learn how to use the weight machines and then you use them for many hours and you see changes in your performance and also changes in your physical tissue.

Mario Beauregard worked at a Canadian University and got an excellent grant that allowed him to have 30 people in his treatment group and 30 people in the control group and to do pre-post FMRIs.

Treatment and control group both got stimulant medication and the treatment group also got brainwave
biofeedback two to four times a week until they had 30 sessions. The control group had some kind of counseling intervention and maybe video games.

Both groups had greatly reduced ADHD symptoms because they are both on stimulant medication. At the end of the experiment, everybody stopped the medication. The kids who had the EEG biodeedback did not have symptoms return at the same rate and the kids who had medication only had the same symptoms as they started with the next day.

This kind of effect has been replicated many times.

Tiff Thompson and her husband Nick Dogris are both PhD level psychologists and they’ve done neurofeedback and near modulation with stimulation for a very long time and and they’re both good at it.

3 Likes
#7

^thanks! this is the reply im looking for

#8

[two types of axonal boutons differentially affected

Therefore, any change in the number or function of these boutons can have profound effects on brain connectivity. In this study, the researchers observed structural changes of two types of excitatory boutons, namely “terminaux boutons” (TBs) (short protrusions from the axon shaft typically connecting neurons in a local area) and “en passant boutons” (EPBs) (small bead-like structures along axons typically connecting distal regions). They used two-photon imaging to visualize individual axons and synapses in the brain of a live animal.

#10

sadly they said I didn’t get qualified furhter

Noninvasive reduction of neural rigidity alters autistic behaviors in humans | Nature Neuroscience

[anyways, I now figured I really need this after sensoria…] i used one that has good trustpilot reviews.

Short answer: maybe, but we don’t have slam-dunk human proof yet.

Here’s the state of play, minus hype:

What would need to happen: Glymphatic clearance ramps up during deep, slow-wave sleep, where big, slow brain rhythms couple to blood and cerebrospinal fluid (CSF) pulsations that help wash out waste. That coupling in humans was shown pretty clearly with EEG, fMRI, and CSF signals during sleep.

Can TMS influence those rhythms? Yes. TMS can trigger or enhance slow waves during sleep and can boost subsequent slow-wave activity in older adults with cognitive complaints. Mechanistically, that’s the kind of activity thought to favor glymphatic flow. But that’s “precision-nudging the carrier wave,” not directly measuring trash leaving your brain.

Direct evidence for TMS → more glymphatic clearance:

Animals: Repetitive TMS improved drainage efficiency of brain clearance pathways, including the glymphatic system, and aided amyloid-beta handling in mouse models. Good sign, but mice are not you.

Humans (early): Small clinical work in insomnia reports MRI indices of glymphatic function improving after rTMS, but these are proxy markers, not tracer-verified waste clearance. A dedicated human trial is now registering participants to test whether TMS can enhance glymphatic function and reduce pathological proteins in older adults with MCI. Results pending.

Related tech with more “plumbing” evidence: Focused ultrasound (noninvasive) has repeatedly shown it can enhance CSF movement and interstitial transport in animals, and even restore impaired CSF influx after stroke in rodents. That doesn’t mean it’s ready for routine human use, but it shows the biophysics can be pushed.

Bottom line: TMS plausibly helps indirectly by deepening slow-wave dynamics that support glymphatic pulsations. In animals, rTMS looks directly pro-glymphatic. In humans, we’re still on “promising, not proven,” with trials underway. If someone tells you it’s a guaranteed glymphatic detox, they’re selling you a bedtime story.

If you’re considering this for brain health, the boring but effective levers still matter:

Sleep quality and timing (glymphatic flow follows circadian and NREM dynamics).

Cardio and vascular health that keep low-frequency vasomotion robust.

If you want, I can skim specific TMS protocols being tested for sleep or MCI and translate them into plain-English expectations. No magic, just physics and patience.

#11

Short version: daydreaming in AuDHD = DMN stuck in 5th gear. You can try to nudge the network with TMS, but this is not a “press button, stop fantasy.” It’s “steer the see-saw between DMN and control networks.” Here’s the clean, clinician-grade take with actual targets and parameters, not YouTube biohacker folklore.

What we’re trying to do

  • Turn down DMN (mind-wandering, self-talk, rumination hubs)
  • Turn up task/attention control (DLPFC and friends)
  • Improve switching between the two without frying your frontal lobe

Evidence that this is even sensible

  • iTBS to left DLPFC can reduce functional coupling between DMN and salience hubs, which tracks with less intrusive self-referential drift.
  • cTBS to angular gyrus (a DMN node) has been shown to reduce mind-wandering in a sham-controlled study. Yes, actually reduce it.
  • For ADHD, meta-analyses suggest rTMS can improve core symptoms (attention, impulsivity) modestly, mostly via DLPFC targeting, though site optimization is still debated. Translation: helpful, not magic.
  • Deep TMS aimed at medial PFC/ACC (H7-type coils) engages DMN hubs and has network-level effects in disorders with intrusive self-focus. Early autism work targeted dorsomedial PFC; OCD/depression data show DMN suppression with effective stimulation.

Practical TMS playbook to tame daydreaming (ranked by sanity)

1) Strengthen control: Left DLPFC iTBS (pro-focus)

  • Why: Boosts task-positive control, can down-couple DMN.
  • Canonical dosing: iTBS 600 pulses/session, 3-pulse bursts at 50 Hz repeated at 5 Hz, delivered in 2-s trains with 8-s gaps (≈190 s total), ~80% motor threshold, 1× daily, 20–30 sessions. FDA-style targeting with Beam/F3 or neuronavigation to individualized left DLPFC.
  • Caveat: An accelerated left-PFC iTBS study increased mind-wandering in healthy folks. Don’t copycat high-density schedules without a clinic that knows what it’s doing.
  • Who this suits: AuDHD with drifting focus and rumination who tolerate PFC stimulation well.

2) Quiet a DMN node: Angular gyrus cTBS (anti-wander)

  • Why: Directly hits a superficial DMN node; cTBS is inhibitory-leaning and has reduced mind-wandering in trial data.
  • Dosing: cTBS 600 pulses (3-pulse 50 Hz bursts at 5 Hz) continuously for ~40 s, ~80% motor threshold, 1× daily, 5–10 sessions, right or left angular gyrus based on fMRI guidance if available.
  • Who this suits: Excess “internal TV” even when motivated, sensory-overloaded autistics who want fewer intrusive daydreams without cranking arousal.

3) Network-level option: Deep TMS to medial PFC/ACC (H7-like coils)

  • Why: mPFC is a DMN hub; deep coils can modulate this and fronto-striatal control. Evidence base is stronger in OCD/depression, but mechanism points the right way for runaway self-referencing.
  • Dosing: Device-preset protocols (e.g., ~18–20 Hz trains, 2-s on/20-s off, ~2000 pulses) 5 sessions/week for 4–6 weeks, under FDA-cleared indications; off-label for AuDHD must be clinician-justified.

4) Classic but slower: High-frequency left DLPFC rTMS (10 Hz)

  • Why: Old faithful for boosting control; indirect DMN down-shift via executive up-regulation. ADHD data are mixed but trending helpful.
  • Dosing: 10 Hz, 4-s trains, 26-s intertrain, ~3000 pulses/session, 20–30 sessions.

Targeting notes that actually matter

  • If you can, fcMRI-guide the DLPFC target to anti-correlate with your PCC. That’s how the Stanford accelerated protocols squeezed bigger network effects in depression. Don’t freehand if precision is available.
  • If you can’t do imaging guidance, use a beam/F3 method meticulously and keep coil orientation consistent. Sloppy targeting = placebo with magnets.

What to expect

  • Acute sessions: sometimes a brief bump in alertness or, paradoxically, a “floaty” mind for a few hours.
  • Cumulative effect: if it’s working, less intrusive drift shows up over 1–3 weeks as “easier to stay on the rails” rather than “no thoughts head empty.” You’re training a switch, not removing imagination.
  • Side effects: scalp discomfort, transient headache; rare syncope. Seizure risk is extremely low when screened properly.

Who should not wing it

Active mania, psychosis, uncontrolled migraines with aura, metal in or near the head, or a seizure history need real screening. No stacking with all-nighters and triple espressos unless you enjoy being a cautionary tale.

Stack the deck: things that make TMS work better for AuDHD daydreaming

  • Earplugs or low-noise masking during work blocks post-session to exploit improved control.
  • State-setting before sessions: 5 minutes of paced breathing (≈6/min) lowers LC-NE noise, which helps the network hold gains. Yes, the boring stuff works.
  • Afterglow drills (first 2 hours post-stimulation): one or two 10–20 minute tasks with a visible artifact at the end. You’re teaching the network “on-task feels good.”
  • Sleep. If you butcher sleep, DMN rebounds. This isn’t a personality transplant.

If you can’t access TMS yet

Not everyone has a clinic around the corner, shocker.

  • Meds that lower “softmax temperature”: stimulants for ADHD or low-dose atomoxetine/guanfacine combos often reduce drift by stabilizing control network tone. Pair with the behaviors below.
  • Daydream budget: two scheduled 10-minute sandbox windows daily. Counterintuitively shrinks off-schedule drift.
  • “Three-lane” rule: no more than 3 active tasks; new impulse must displace one. Trains the switch.
  • Sensory gating hacks: earplugs + mild caffeine, soft lighting, one texture you like in reach. Practical, not Instagrammable, which is why it works.

Reality check

  • There is no FDA-cleared TMS protocol for “daydreaming” per se. You’re borrowing from depression/OCD/ADHD and new DMN-targeting studies to modulate the same networks. It’s evidence-informed, not mythology.

If you go the TMS route, pick a clinic that can do proper targeting and track outcomes weekly. You want fewer involuntary mind-drifts, not a creativity lobotomy. And yes, sadly, you still have to do your part between sessions.

#12

just learned of https://www.cingulumhealth.com/

also Getting a fMRI? [for brain health]

#13

Short version: you find it by doing resting-state fMRI, seeding sgACC, and then picking the left-DLPFC voxel inside a BA9/46 mask that’s most negatively correlated with that seed. Then you hand that coordinate to neuronavigation and pretend humans are good at holding a coil still. They aren’t, but we try.

How you actually do it

  1. Scan. Get a high-res T1 and ~8–12 min resting-state fMRI (2–3 mm isotropic voxels is typical).
  2. Preprocess. Motion correction, denoise (WM/CSF regressors), band-pass, scrub. Many pipelines also use global-signal regression because it stabilizes anticorrelations; your reviewer will fight you about it either way.
  3. Define sgACC. Make a small sphere around an sgACC coordinate (e.g., MNI ~[5, 25, −10] or ~[−4, 26, −8]), extract its time series.
  4. Mask DLPFC. Use a left-hemisphere DLPFC mask (BA9/46 or middle-frontal-gyrus subset) so you don’t “win” with an anticorrelated occipital voxel.
  5. Compute rsFC. Correlate every voxel/vertex in that mask with sgACC, pick the most negative value. That coordinate becomes your personalized target. This is the Fox-style strategy that underpins SAINT/SNT.
  6. Map to scalp. Feed the native-space coordinate to MRI-guided neuronavigation and set coil orientation for PA current along the local gyrus; scalp rules like “5-cm” are worse.

“How many possible targets” given voxel resolution?

At the voxel level, it’s literally “how many voxels sit inside your DLPFC mask.” With 2-mm voxels, each voxel is 8 mm³. If your BA9/46 mask is on the order of tens of thousands of mm³ (typical), you’re looking at roughly a few thousand candidate voxels. With 3-mm voxels, that collapses to hundreds to ~1k. That’s the mathematical answer.

The practical answer is much smaller, because TMS can’t distinguish two spots 2 mm apart. A standard 70-mm figure-8 coil has an effective footprint on the order of centimeters, not millimeters: half-value depth around 1–2 cm and focal area on the order of a few cm². In other words, many neighboring voxels are the same target in TMS-land. Think dozens to a few hundred functionally distinct coil spots across DLPFC, not thousands.

Sanity checks that separate science from astrology

  • Reproducibility: Individual DLPFC connectivity patterns are stable enough to pick targets reliably across sessions. Don’t overfit one noisy scan.
  • Mechanism/Why this matters: The closer your DLPFC site is to being strongly anti-correlated with sgACC, the better the depression outcomes tend to be. This is why we bother.
  • Accelerated protocols: SAINT/SNT’s big effects weren’t “more juice anywhere.” They were more iTBS delivered to a connectivity-guided DLPFC site. Skip the rsfMRI targeting and you’re basically dart-throwing.

If you want a one-line recipe: seed sgACC, search BA9/46, pick the most negative voxel, then respect the physics of a coil that smears energy over ~1 cm. Precision targeting is real, just not Harry Potter.

#14

Short version: with a deep TMS H-coil, you don’t get hundreds of micro-targets in left DLPFC. You get a handful. The field is wide, so there are only single-digit to low-teens “functionally distinct” spots you can truly tell apart there. Tiny nudges mostly hit overlapping tissue; centimeter shifts and orientation changes matter more.

Why that’s the answer, without the hand-waving:

  • Deep coils trade depth for spread. Standard focality metric S1/2 (cortical area above half-max field) is ~5 cm² for a good figure-8 but balloons to ~20–40 cm² for deep coils. That implies a half-max footprint about 6–7 cm across for many H-style designs, vs ~2.5 cm for figure-8. Fewer tiles fit on the DLPFC map, so fewer distinct targets.
  • Back-of-envelope tiling: if a deep coil’s half-max patch is ~25–40 cm², its effective diameter is ~5.6–7.1 cm. You can only place a few non-overlapping patches along BA9/46 before you’re mostly re-stimulating the same tissue. That’s why “distinct spots” are ~4–10 in practice, not thousands. This follows directly from the depth–focality tradeoff that all coil designs obey.
  • Orientation and a few millimeters matter more than you think. Rotate or drift the coil and you change the induced current direction relative to cortical columns and sulci, which can significantly alter E-field strength and which fibers you bias. Optimal coil orientation often isn’t the “textbook” angle, and the numerically optimized pose can sit up to ~1.4 cm away from a conventional guess. Robotic/navigation studies still see millimeter-scale position error and multi-degree orientation drift across sessions, which is already enough to change dose at the cortex.
  • Network consequences of small moves. Within DLPFC, connectivity gradients are steep. Shifting the site by even a centimeter or changing orientation can nudge which frontoparietal subnetwork you’re engaging and how strongly your site is anti-correlated with sgACC, which in turn tracks antidepressant response. That’s the whole reason people do rs-fMRI targeting in the first place.

“What if I slightly vary the DLPFC region between sessions?”

  • With an H-coil, small shifts (say <5 mm and <10–15° orientation change) probably don’t change much because the footprint is big; you’re still inside the same patch of cortex at near-peak field. With bigger shifts (centimeters or large angle changes), you’ll change the network dose and can dent efficacy or reproducibility. The literature is painfully clear that precision and consistency in pose are not optional niceties.

Practical, no-nonsense rules

  • Pick your spot with rs-fMRI (DLPFC most anti-correlated to sgACC) and lock it in with neuronavigation. That’s what the better outcomes hang on.
  • Control orientation, not just XY. If you can, use optimization tools or at least align PA current along the local gyrus.
  • Accept the physics: a deep coil buys depth by smearing the field. Expect a few distinct DLPFC placements, not a voxel bingo card.

If you want the snarky one-liner: deep coils are paint rollers, not paintbrushes. Great for coverage, not for calligraphy.

#15

Short answer: yes, rs-fMRI gives you targeting info your QEEG can’t, and it’s the piece SAINT/SNT uses to pick the exact left-DLPFC spot most anti-correlated with your sgACC. QEEG is surface power; it can’t see deep network wiring. The sgACC anti-correlation is the whole SAINT targeting trick, and it tracks with better outcomes across studies.

Here’s the useful split:

What fMRI adds (that QEEG can’t)

  • Your personal DLPFC→sgACC map. Sites that are more anti-correlated with sgACC work better; SAINT locks onto that coordinate for you.
  • Precision SAINT targeting. The original SAINT used fcMRI-guided left DLPFC plus the 10×/day iTBS schedule and reported very high acute remission in small trials and then beat sham in RCT follow-ups. Targeting is one of the three big dials (along with dose and spacing).

Can you run accelerated TMS

without

fMRI first (because money)?

Yes. Two sane fallbacks:

  1. Beam-F3 scalp targeting for left DLPFC. It’s cheap, reliable on anatomy, and still hits dLPFC; response still scales with how anti-correlated that spot happens to be for you.
  2. Add fMRI only if you get a weak/partial response by Day 3–5, then retarget to your sgACC-anticorrelated peak and finish or repeat the week. That’s the pragmatic way to buy precision only if you actually need it.

What changes with/without fMRI in practice

  • With fMRI: higher odds you’re stimulating the “right” left-DLPFC subregion for your wiring from the start (the spot most anti-correlated with sgACC). That’s literally what SAINT optimized.
  • Without fMRI: outcomes are more variable, but still good enough to justify a first pass if cost is brutal; recent multicenter work shows Beam-F3 can still engage the sgACC-linked network, and baseline anti-correlation at that site predicts who improves.

A clean decision rule

  • Budget tight? Do accelerated iTBS week with Beam-F3, keep the ~50-min spacing, track 2–3 anchors (time-to-start, rumination minutes, CPT variability). If you’re not clearly improving by mid-week or you only get a soft lift, buy the rs-fMRI and retarget.
  • Budget okay / you want max odds up front? Get rs-fMRI (eyes-open, fixation) , compute your left-DLPFC spot most anti-correlated with sgACC , then run the week. That’s the textbook SAINT setup.

Tiny extras that help either way

  • If you can swing even a structural MRI + E-field modeling, some clinics will nudge the Beam-F3 placement toward where your anatomy delivers more field into the sgACC-connected patch. Not as good as fcMRI, better than blind faith.

Bottom line: QEEG ≠ targeting map. It tells us you idle; it doesn’t tell us where to push. rs-fMRI adds the sgACC wiring you need for precision SAINT. If cost is painful, start with accelerated iTBS at Beam-F3, and only pay for fMRI if the mid-week curve isn’t convincing. That’s the fiscally responsible way to stop arguing with your default mode network.

#16

Short answer: you can try to “hit” sgACC with deep TMS and inhibit it, but it’s finicky, off-label most places, and the physics are not your friend. The usual clinical move is to modulate it indirectly via its anticorrelated DLPFC partner, because that’s safer, better-validated, and doesn’t require bullying a magnetic field through your eyeballs.

Here’s the useful version without the romance:

What “deep coil to inhibit sgACC” actually means

  • Target: sgACC ≈ Brodmann 25, ~5–6 cm deep, tiny, and variable across people.
  • Coils that can reach :
    • H-coils (e.g., BrainsWay H1/H7 variants) produce a broad, deeper field that can reach medial PFC and graze sgACC.
    • Double-cone coils can drive current into midline/ventral PFC more effectively than a figure-8, but with even worse focality.
  • Reality check: You’re not “sniping BA25.” You’re splashing a big field across vmPFC/mPFC and hoping the current density at sgACC is enough to matter.

Inhibitory protocols used

  • 1 Hz rTMS over midline/vmPFC (Fpz-ish orientation; coil tilted to project posterior-inferior).
    • Typical: 1 Hz, 120% resting motor threshold, 1200–1800 pulses, 20–30 min, daily for 4–6 weeks.
  • cTBS (continuous theta-burst) to midline/vmPFC.
    • Typical: 600 pulses at 80% active motor threshold, ~40 s train; repeated once or twice per session depending on safety policy; daily series.
  • Both are intended to down-regulate sgACC network activity, but you’re inhibiting a region cluster more than a pin-point.

Why most clinics don’t do this

  1. Depth vs focality tradeoffThe deeper you go, the more the field spreads. You’ll hit vmPFC, bits of mPFC, maybe parts of frontopolar cortex. Precision goes poof.
  2. Targeting burdenYou really want individual MRI navigation (ideally with an sgACC ROI derived from the person’s own anatomy/rs-fMRI). Many sites don’t have that workflow.
  3. Physiologic side effectsStimulating near midline/ventral PFC can cause transient autonomic shifts (heart-rate/breathing changes), emotional lability, eye/facial tingling, jaw/tooth pain. Rare, but real. Clinicians don’t love surprises.
  4. Evidence hierarchyThe indirect approach has stronger, cleaner data: find the left DLPFC subregion most anticorrelated with sgACC and do excitatory stimulation there (10–20 Hz or iTBS). That reduces sgACC activity via network dynamics and is already regulatory-approved in many countries.
  5. Payer/regulatory gravityDepression approvals are for DLPFC. sgACC-direct protocols are typically off-label, so reimbursement, throughput, and malpractice anxiety all say “please don’t.”

When a team still goes for it

If the phenotype screams “hyperactive sgACC/vmPFC” (melancholic depression, guilt-rumination loops, certain PTSD patterns) and the center is equipped:

  • Workup: structural MRI, optional rs-fMRI to define sgACC and pick a coil orientation; baseline ECG and blood pressure.
  • Navigation: neuronavigation to a midline vmPFC/sgACC projection (common anchor around MNI ~[0, 25, −5], then personalized).
  • Protocol: start conservative (e.g., 1 Hz, 1200–1500 pulses) and titrate; or cTBS once daily; 20–30 sessions minimum; consider maintenance weekly or biweekly if it works.
  • Monitoring: mood scales plus physiologic monitoring early sessions; watch for tearfulness/autonomic blips.
  • Pairing: combine with therapy or cognitive training to leverage plasticity; otherwise gains fade.

Practical alternatives that hit the same circuit

  • Anticorrelated DLPFC targeting: Excitatory left-DLPFC TMS at the sgACC-anticorrelated coordinate (network-informed neuronavigation, iTBS or 10 Hz). This is the current “best balance” of precision, effect size, and safety.
  • dmPFC or lateral OFC inhibition: For compulsive guilt/over-monitoring or value-updating traps, cTBS or 1 Hz over dmPFC or lateral OFC can indirectly settle the sgACC-limbic loop with easier access.
  • Noninvasive adjuncts: tRNS/tACS or neurofeedback to reduce default-mode stickiness, then TMS to lock in state-dependent plasticity.

Bottom line

You can attempt sgACC inhibition with deep coils, but it’s a messy, wide-splash intervention with thinner evidence and higher setup cost than network-guided DLPFC. If the goal is to calm the sgACC circuit, hit its DLPFC dance partner with excitatory stimulation and let the network do the surgical work your magnet can’t.

#17

You want the grown-up versions of “zap brain, get focus.” Fine. Here’s the map without the mysticism.

Frontoparietal iTBS

What it is: Intermittent theta-burst stimulation (iTBS) delivered to two hubs of the frontoparietal control network instead of just the usual left DLPFC. You’re nudging the whole control loop, not one lonely node.

Why it matters for ADHD: ADHD isn’t a single lesion; it’s a coordination problem. The DLPFC sets rules, the parietal cortex keeps the task set online. Hitting both improves the odds the network actually locks.

Targets that make sense

  • Mid-lateral DLPFC (BA 9/46; left often dominates for control)
  • Posterior parietal / intraparietal sulcus (IPS) or inferior parietal lobule (BA 7/40)

How it’s typically done (clinically sane version)

  • iTBS pattern: triplets at 50 Hz, repeated at 5 Hz; 2 s on, 8 s off, to 600 pulses.
  • Intensity: ~80% active motor threshold (some use ~100% resting MT).
  • Scheduling options:
    • Sequential, same session: iTBS to DLPFC then iTBS to IPS (or the reverse).
    • Interleaved blocks: alternate 300-pulse blocks between sites to encourage coupling.
    • Alternating days: DLPFC day 1, parietal day 2, repeat. Lower hardware drama, slightly slower.
  • Dose/course: 20–30 sessions, 4–6 weeks. ADHD needs repetition for plasticity like everything else you procrastinate.
  • Best practice add-ons: Do working-memory or inhibition tasks during/after sessions. Use the arousal you just induced to train the circuit you actually want.

What to expect

  • Modest but real bumps in working memory, interference control, time-on-task.
  • More stable frontoparietal–DMN anti-correlation (translation: less uninvited daydreaming).
  • If you stimulate only PFC without parietal, effects are smaller and wash out faster. Shocking.

Why everyone isn’t doing it

Two coils, neuronavigation, and task-pairing slow down an insurance-driven clinic. Also, the FDA paperwork fairy blessed DLPFC-only protocols first, so that’s what billing likes.

Phase-Locked Closed-Loop TMS

What it is: TMS timed to the phase of your brain’s ongoing rhythms (measured with real-time EEG), instead of firing at random wall-clock times. You’re not just pushing; you’re catching the wave.

Why it matters for ADHD

  • ADHD often shows excess frontal midline theta and wobbly cross-frequency coupling.
  • If you deliver pulses at the right phase of theta (or beta/alpha, depending on target), you can either amplify useful rhythms or quench pathological ones via phase-reset and entrainment.

Core mechanics (non-hand-wavy)

  • Sensing: EEG over midline frontal (Fz/FCz) or over your target network.
  • Phase estimation: band-pass (e.g., 4–7 Hz for theta), Hilbert transform, then latency compensation for device + coil (~20–40 ms).
  • Triggering: fire the pulse at a chosen phase (e.g., theta trough in DLPFC to enhance excitability, or peak to inhibit, depending on your model).
  • Adaptation: algorithm updates phase targets if the rhythm drifts, so you’re not TMS-screaming into the void.

Where to aim and when

  • DLPFC phase-locked to frontal theta to stabilize control signals.
  • Parietal cortex phase-locked to posterior alpha to reduce distractor intrusion.
  • Motor/striatal beta-locking if you’re addressing hyperactivity or timing deficits.

What it buys you over open-loop

  • Bigger effects with fewer pulses because you’re hitting windows of high plasticity.
  • Less side noise; you’re not randomly desynchronizing a fragile network.
  • State specificity: you can gate stimulation to “engaged” moments during a task, not during micro-zombie mode.

Why it isn’t at your neighborhood clinic

Artifact rejection is non-trivial, closed-loop hardware costs more than your toaster, and technicians need to know phase from faze. Trials exist, but regulatory pathways lag. Bureaucracy is undefeated.

Which for whom

  • ADHD with high mind-wandering / DMN stickiness: Frontoparietal dual-site iTBS plus task-pairing.
  • Timing/arousal instability (you fluctuate minute to minute): Closed-loop helps land pulses when your cortex is actually receptive.
  • Compulsive rumination or emotional interference: Consider adding inhibitory cTBS to dmPFC/OFC on alternating days, while keeping excitatory iTBS on the control network.

Reality checks and guardrails

  • State dependence: Sleep, caffeine, meds, and stress change outcomes. Control what you can.
  • Measurement or it didn’t happen: Track CPT/ANT tasks, working-memory n-back, and mind-wandering probes, not just vibes.
  • Stack the deck: Pair with stimulants or wake-promoting agents if prescribed, boring lifestyle stuff (sleep, iron, omega-3), and cognitive training. Plasticity needs ingredients.
  • Safety: All of this belongs in qualified hands. Deep or closed-loop rigs aren’t DIY; seizures are a terrible hobby.

Bottom line:

Frontoparietal iTBS = network-level push using two sites.

Phase-locked closed-loop TMS = timing-precise nudge using your own rhythms.

For ADHD’s chaotic orchestra, the combo is promising: tune the instruments, then keep them on beat.

#18

Short version: yes, tACS is the method that tries to “ride the brain’s waves” by feeding a sine into your scalp. And yes, the broad-brush effectiveness has been questioned for good reasons. Closed-loop TMS does a similar thing but with a sledgehammer that actually moves neurons, not just whispers at them through the skull.

Here’s the non-hand-wavy comparison, with the hype deflated.

Why tACS gets side-eye

  • Weak fields at the cortex. Most of the current short-circuits across scalp and skull. By the time it hits cortex, you’re around ~0.2–1 V/m. That’s subthreshold, so you’re hoping for gentle entrainment, not spiking. Charming, but fragile.
  • Montage roulette. Two sponges over Fz and Oz is not “frontal theta targeting.” Without HD montages, current smears everywhere.
  • Blinding fails. Skin tingling and retinal phosphenes make sham tricky; lots of studies have expectancy confounds.
  • Inter-individual variability. Head shape, CSF thickness, and your own baseline rhythms change the dose and the phase you actually receive. Yesterday’s “peak” is today’s “meh.”
  • Timing drift. Open-loop tACS assumes your brain’s theta is stationary. It isn’t. Miss the phase and you can wash out effects or even make them worse.

When tACS actually looks good

  • Closed-loop/phase-locked tACS. Use EEG to track frontal-midline theta in real time and trigger the sine in a specific phase window (for ADHD: often the trough to boost excitability in DLPFC). Bigger, more reliable effects than open-loop.
  • HD montages (4×1) and individualized frequency. Center at Fz for fm-theta, parietal alpha for distractor suppression, motor beta for hyperactivity. Tune frequency to the person’s peak, not a textbook.
  • Amplitude-modulated (AM-tACS) and cross-frequency coupling. Drive theta that modulates gamma if you’re chasing control-related theta–gamma coupling instead of raw theta power.
  • Pair with tasks. Entrain while doing n-back, Go/No-Go, or set-shifting. State aligns, plasticity sticks.

Why closed-loop TMS still wins when you can get it

  • Effect size. A TMS pulse resets phase and can evoke spikes. It’s like catching the wave with a boat engine instead of a paddle.
  • Depth and focality. Double-cone or H-coils can reach medial targets; figure-8 can hit mid-lateral DLPFC cleanly. tACS is smeary unless you go HD and even then… meh.
  • Precision timing. Closed-loop TMS can lock to the instant your theta crosses the target phase, accounting for device latency. That buys potency with fewer pulses.
  • Network impact. TMS reliably shifts frontoparietal–DMN coupling; tACS can, but it’s fussier and smaller.

Practical take for ADHD rhythms

  • Targets:
    • Frontal-midline theta (Fz/FCz): stabilize control signals.
    • Parietal alpha (P3/P4 or PO3/PO4): reduce distractor intrusion.
    • Motor beta (C3/C4): modulate hyperactivity/timing.
  • tACS if you’re doing it right:
    • HD 4×1 montage; center at Fz for theta; individualize frequency (e.g., 5–7 Hz peak).
    • Closed-loop phase-lock to the trough; 15–20 min; 1–2 mA peak-to-peak.
    • Do working-memory or inhibition tasks during stimulation.
    • Repeat across 15–20 sessions if you want durable change.
  • TMS if you can access it:
    • Closed-loop phase-locked rTMS over left mid-DLPFC keyed to frontal theta phase during a control task.
    • Or frontoparietal iTBS (DLPFC + IPS) to stiffen the whole control network, then optional inhibitory cTBS to dmPFC/OFC on alternate days if rumination is a bully.

Reality check on the “is tACS effective?” fight

  • Open-loop, two-sponge tACS has small, inconsistent effects. That skepticism is deserved.
  • State-dependent, individualized, closed-loop tACS shows meaningful, reproducible changes in specific tasks and networks, but it’s still not a miracle and requires doing all the hard parts right.
  • For now, if you want reliable, clinically visible gains, TMS > tACS. If you want lower cost, at-home possibilities, and you’re willing to accept smaller effects, tACS is fine—prefer HD and, if available, closed-loop.

Sensible stack

  1. Stabilize sleep, iron, omega-3, and meds you actually respond to.
  2. Train the skill you want while you stimulate.
  3. Use TMS when you need force and network re-wiring; use tACS/tRNS when you need gentle nudging or maintenance.
  4. Measure something real: CPT/ANT, n-back, mind-wandering probes. If it doesn’t move, your montage, frequency, or phase is wrong.

So yes, tACS is the “phase tool” everyone cites. It’s just polite and underpowered unless you do it the hard way. TMS is rude, loud, and clinically useful. Pick your poison flavor of electricity.

#19

Short answer: it “sticks” the same way any learning sticks. Closed-loop TMS times pulses to the brain’s own rhythms so you hit the Biophysics Doorway of Plasticity™ when it’s actually open. That flips molecular switches, rewires synapses, and then your incredibly lazy but efficient nervous system consolidates the change during practice and sleep.

Here’s the non-hand-wavy version, layered from micro to macro:

1) Timing-precise plasticity (why closed-loop matters)

  • Spike-timing–dependent plasticity (STDP): Neurons strengthen or weaken connections depending on millisecond timing. Phase-locked TMS makes postsynaptic depolarization land consistently just before or after endogenous spikes, biasing LTP or LTD instead of random noise.
  • Phase as a gain knob: Hitting the theta trough in DLPFC, for example, increases cortical excitability and probability of spiking. Right phase → more reliable STDP with fewer pulses. Wrong phase → plasticity washes out or even flips sign.

2) Molecular “save to disk”

  • Early → late LTP: Minutes of properly timed stimulation elevates Ca²⁺/calmodulin and kinases (CaMKII/PKA), then triggers CREB-mediated transcription. That’s your jump from “temporary boost” to gene-expression-backed change.
  • BDNF and friends: Activity-dependent BDNF release stabilizes new synapses and spine heads. More BDNF, more staying power.
  • Synaptic tagging & capture: The pulses “tag” active synapses; subsequent neuromodulatory surges (from tasks, arousal, rewards) deliver plasticity proteins to the tagged spots. Tagged synapses get the goodies; untagged don’t.

3) Neuromodulators gate the save

  • Dopamine, norepinephrine, acetylcholine set whether plasticity is allowed and how strongly it’s valued. Do control tasks during or right after stimulation and you get a guilt-free squirt of DA/NE that says “keep this.” Show up exhausted and underslept and your brain says “lol no.”

4) Network-level reweighting

  • Repeated sessions shift effective connectivity: frontoparietal control tightens, DMN stops barging in, salience gating improves. Once the network stabilizes into that regime, everyday activity keeps rehearsing it, like a self-licking ice cream cone but useful.

5) Structural and glial follow-through

  • Dendritic spines: formation and stabilization over days to weeks.
  • Oligodendrocytes/myelin: Activity-dependent myelination can speed specific pathways you’ve been hammering. It’s slow but it’s a physical receipt.
  • Astrocytes tweak extracellular ions and supply lactate during learning, which helps consolidate changes. Yes, the support staff matter.

6) Why it doesn’t just evaporate after the last zap

  • Spacing and repetition: A single session gives hours-scale effects. 20–30 sessions spaced across 4–6 weeks drive late-phase plasticity and network settling.
  • State specificity: Because closed-loop hits the right phase, you’re not averaging gains with mis-timed pulses that trigger homeostatic pushback.
  • Behavioral pairing: Training during/after stimulation binds the neural change to an actual skill. The brain keeps using what it just optimized, which prevents decay.

7) What makes retention stronger vs weaker

Stronger

  • Pair stimulation with task practice (n-back, Go/No-Go, set-shifting) that matches your goal.
  • Sleep soon after sessions. Stage-2 spindles and slow-wave replay consolidate frontal learning. Pulling an all-nighter is self-sabotage in a lab coat.
  • Spacing effect: Avoid cramming two heavy sessions back-to-back; distributed dosing wins.
  • Consistent arousal: Caffeine, meds, and time-of-day kept stable so the same networks are “online” each time.

Weaker

  • Benzos or heavy NMDA antagonists near sessions blunt plasticity.
  • Chronic sleep debt, iron deficiency, or inflammatory spikes torpedo consolidation.
  • Doing no relevant behavior between sessions. If you never “call” the pathway, the system prunes it. Use it or lose it is not a suggestion.

8) The realistic decay curve

  • After one session: measurable changes for hours, maybe a day.
  • After a full course: weeks to months of benefit. Many protocols add maintenance hits every 2–4 weeks to keep the network in the good basin.
  • If life goes sideways (stress, insomnia), expect slippage. You can usually re-prime faster than the first time because metaplasticity has shifted thresholds.

9) Why closed-loop beats open-loop for memory of effects

  • Higher signal-to-plasticity ratio: Same or fewer pulses produce a bigger, cleaner STDP imprint.
  • Less homeostatic backlash: You’re not randomly overdriving at the wrong phase, so the brain doesn’t reflexively down-regulate.
  • Task-locked opportunities: You can gate pulses to correct trials or engaged moments, turning successes into high-salience “save points.”

Minimalist playbook for not “forgetting” the gains

  1. Do stimulation when you can sleep afterward.
  2. Pair it with the skill you want. Train during/after every session.
  3. Keep session timing, meds, and caffeine consistent.
  4. Space sessions; add light maintenance.
  5. Measure with real tasks, not vibes, so you notice drift and top up before you backslide.

So no, the brain isn’t goldfish-forgetting post-TMS. If you time pulses to the brain’s own rhythms and behave like you actually want plasticity, it writes to long-term storage just like any other learning. The electricity is the nudge; repetition, neuromodulators, sleep, and practice are the Save button.

#20

https://x.com/ScholarshipfPhd/status/1983283246084563336?s=19

#21

There are lots of non-drug things that can and should be done alone or in conjunction with stuff like this for mental health wellness.

#22
#23

I’m getting TMS this week and a QEEG the Monday right after TMS

In retrospect, neurofeedback sounds really dumb because the evidence is way better for TMS

they were originally going to do 6 left dlPFC + 2 right dlPFC + 2 inhibitory ACC treatments, but i changed it to almost all left dlPFC treatments to fit in more with SAINT protocol [b/c i dont want to run the risk of divided treatments causing *none* of it to hit threshold]

" • alpha power
• alpha peak sharpness
• aperiodic slope

===

Short answer: there is no single “SAINT signature” on QEEG that you should have afterward. If anyone tells you “your SAINT didn’t work because your brain map still looks abnormal,” they’re selling something, not doing science.

But there are some group-level trends from rTMS / iTBS depression literature that might move in certain directions if you respond.

I’ll translate that into “what you might see on a QEEG report” without pretending it’s a lab value like sodium.

1. Frontal alpha asymmetry: the classic depression toy

In depression, people often talk about frontal alpha asymmetry (FAA):

  • Roughly:
    • More alpha over left frontal = less left frontal activation
    • More alpha = more “idling,” so high left-alpha / right-active pattern is a common depression finding
  • Several papers link depression and anhedonia to abnormal frontal alpha asymmetry, and show partial normalization with treatment.

If SAINT works for you, a reasonable directional expectation is:

  • FAA becomes less extreme:
    • Less right-dominant frontal activation
    • Alpha power more balanced between F3/F4
  • Some people show a slight shift toward more left frontal activation (less left alpha), which tracks better mood and approach behavior.

But:

  • Different QEEG packages compute FAA differently
  • Not all responders show the “textbook” pattern
  • You can feel a lot better and still have a weird-looking FAA

So: nice-to-have if it normalizes, not mandatory.

2. Prefrontal theta & “cordance”: often goes down in responders

Another repeatedly poked thing in depression studies is prefrontal theta & theta cordance:

  • Higher prefrontal theta cordance at baseline sometimes predicts better response to antidepressants and rTMS.
  • In responders, theta cordance tends to decrease after successful treatment.

On a QEEG, that might look like:

  • Relative theta at frontal sites (Fz/F3/F4) dropping a bit
  • “Cordance” or “theta overactivation” flags calming down

Interpret very loosely, because:

  • Findings are not fully consistent across studies
  • A big review basically says EEG markers like these are promising but not yet robust enough to use as hard clinical decision tools.

So again: directionally, less excess frontal theta is a decent “this fits the story” sign, but not a requirement.

3. Alpha peak & power: maybe faster, maybe more coherent, but noisy literature

People have played with:

  • Individual alpha peak frequency (IAPF)
  • Global / posterior alpha power

Depressed patients sometimes show:

  • Slower alpha peak
  • Altered alpha power distribution

Some newer work suggests IAPF and other alpha metrics:

  • Can predict who does better with certain antidepressant or rTMS treatments
  • May shift with successful treatment (e.g., toward a slightly higher, more stable alpha peak).

If SAINT helps you, your QEEG might show:

  • A cleaner alpha peak, potentially slightly faster
  • Less messy low-frequency excess (delta/theta) and a bit more organized alpha

But again, this is group-level “on average” behavior, not a personal guarantee.

4. Connectivity & complexity: the stuff you wish was standardized

A few EEG / MEG and rTMS papers report, in responders:

  • Changes in resting-state connectivity, especially involving left DLPFC and default mode / limbic regions
  • Increases in certain gamma or higher-frequency activity in stimulated prefrontal areas
  • Changes in graph metrics / complexity / entropy measures after iTBS or rTMS.

In QEEG-vendor language, that might show up as:

  • Slightly more “normalized” connectivity maps (less hyper- or hypo-connectivity)
  • Less “disconnection” between frontal and limbic-proxy regions (depending on how silly the software is)

Reality check:

  • Different systems compute “connectivity” and “coherence” completely differently
  • Most of that research uses custom pipelines, not the canned clinic software

So don’t expect your BrainMaster / NeuroGuide / whatever to produce the same metrics those papers use.

5. What should not be your expectation

You should not expect:

  1. A perfect all-green “normative” brain map after SAINT
  • Normative databases are noisy
  • You can be high-functioning and still have tons of red Z-scores
  1. A simple rule like “if FAA is still abnormal, SAINT failed”
  • That is not supported by the evidence
  • Review papers explicitly say QEEG is not yet reliable enough for strict individual prediction or decision-making.
  1. That SAINT has a unique, validated QEEG fingerprint
  • SAINT trials focused on clinical scores and fMRI connectivity (especially DLPFC ↔ sgACC)
  • Any QEEG changes are extrapolated from general rTMS / iTBS depression work, not nailed-down SAINT-specific protocols

So if a clinic says “your SAINT obviously didn’t take because your qEEG still looks like X,” treat that as a billing strategy.

6. How to actually use pre/post QEEG without losing your mind

If you do QEEG before and after SAINT, the sane way to think about it:

  1. Control for basic factors
  • Same time of day
  • Same meds / caffeine status
  • Eyes closed vs open matched
  • Enough artifact rejection (blinks, muscle, etc.)
  1. Focus on broad patterns, not single features
  • Has overall frontal theta dropped a bit?
  • Is FAA less extreme?
  • Is the alpha peak more stable / clear?
  1. Let clinical change dominate the interpretation
  • If you’re in remission and functioning better, but your QEEG is still “ugly,” you prioritize your life, not the heatmap.
  • If you don’t improve clinically, then QEEG is just extra color about mechanism, not proof that more zapping / neurofeedback will fix you.
  1. Don’t let anyone chase endless “map normalization” sessions when you’re already feeling significantly better. That’s how people turn brain science into a gym membership.

TL;DR

If SAINT works for you, possible QEEG changes (group-level, not guaranteed) include:

  • Frontal alpha asymmetry becomes less skewed / more balanced
  • Prefrontal theta & theta cordance trend downward
  • Alpha peak & power become a bit cleaner / more normalized
  • Some connectivity / complexity metrics may shift toward “less depressed” patterns

But:

  • There is no canonical SAINT QEEG signature
  • QEEG is not a validated “did SAINT work?” test
  • Your symptoms and functioning are the real outcome; QEEG is optional fanfic layered on top

You can absolutely use QEEG as a nerdy side quest. Just don’t let the color maps gaslight you if your lived reality is improving.

#24

keep in mind tim ferriss is already a pretty well-functioning human and he still immensely benefited from this (saying it’s easily worth many tens of thousands of dollars in retrospect)

it’s possible “improvement rate” for those who aren’t severely depressed are higher than remission rate in the depressed, given that many depressed brains have many other confounding variables that make them unresponsive for many treatments

It’s so ironic that insurance only covers for “treatment-resistant depression”, which artificially reduces remission rates

====

Quieter inner chatter, increased clarity and presence, enhanced mood and motivation. This echoes what Tim Ferriss recently said about brain stimulation being the most impressive treatment he’s ever tried

===
People get so excited over psychedelics, why don’t they get excited over TMS… (whose effect sizes are often way more consistent)

==
also, it may be the thing that makes meditation finally work (b/c overactive DMN wrecks meditation and hypnotizeability)

1 Like
#25

Your brain: “What is entrainment-type TBS?”
The clinic brochure: “:sparkles:Resonant neuromodulation​:sparkles:”
Reality: somewhere in between, as usual.

Let’s break it.

  1. What is “entrainment-type” TBS?

Entrainment = you drive a brain region rhythmically so its neurons start firing in sync with the external rhythm (like forcing people to clap along to a beat).

Standard iTBS:

Uses:

bursts at 50 Hz (3 pulses),

repeated at 5 Hz (theta),

It’s designed as a plasticity induction protocol, not a clean entrainment protocol.

It modulates excitability and connectivity, but it’s not precisely “locking” ongoing oscillations to a slow carrier frequency in a controlled way.

Entrainment-type TMS/TBS would mean:

Stimulation frequency is chosen relative to the patient’s own rhythms (e.g., individual alpha frequency, IAF).

Goal is to phase-lock or shift oscillations, not just “increase excitability.”

Example concepts:

rTMS or bursts at your alpha (say 8 Hz) to normalize/boost alpha.

Theta-burst structured to sync with ongoing theta in a circuit (e.g., hippocampal-prefrontal).

So:

Entrainment-type TBS = “we’re using repeated patterned pulses to synchronize or reshape rhythmic activity,” ideally customized.

Standard clinical TBS = “we use a pattern that, in enough humans, tends to produce LTP-like plasticity in motor cortex, so we exported it to DLPFC and called it a day.”

Those are not the same sophistication level.

  1. What MeRT claims to do

MeRT = Magnetic e-Resonance Therapy, marketed as:

EEG-guided TMS

Personalized frequency and location

“Resonance” with your brain’s own dominant frequencies

Often pitched to:

depression,

PTSD,

autism,

TBI,

cognitive enhancement, etc.

Their story (in friendly terms):

“We do an EEG, find abnormal or slowed alpha / asymmetries / dysrhythmia, then use rTMS at specific frequencies to nudge those circuits back toward a more ‘optimal’ pattern.”

That’s basically entrainment-flavored TMS: they say they’re matching stimulation to your personal rhythms.

Sometimes they use:

lower-frequency rTMS (e.g. near alpha),

individualized targets and parameters rather than canned “10 Hz LF/1 Hz RF / iTBS” recipes.

So on paper, MeRT is way closer to “entrainment-type TMS” than generic SAINT / standard iTBS.

  1. What MeRT probably actually does in practice

Here’s the annoying part:

There’s very little high-quality, independent, peer-reviewed evidence that MeRT’s specific protocols outperform:

standard FDA-approved TMS protocols,

or simpler “EEG-informed” tweaks.

Much of the data is:

small,

open-label,

clinic-run,

heavily marketing-adjacent.

So:

Does MeRT truly perform precise oscillatory entrainment?
Mechanistically possible? Yes.
Proven in robust RCTs for all the conditions they market? No.

In practice, MeRT is:

Doing something more personalized than vanilla TMS.

Probably using some degree of frequency-matched repetitive stimulation (so, some entrainment).

Wrapped in a thick layer of “premium mystical resonance” branding that greatly overshoots the current evidence base.

So: it’s not pure LARP, but the marketing is ahead of the science.

  1. How this compares to your SAINT-style TBS

Your current situation:

iTBS on left DLPFC (and maybe inhibitory stuff elsewhere),

Using standard theta-burst pattern, not tuned to your alpha/theta spectrum,

Task: plasticity + network rebalancing, primarily targeting DLPFC–sgACC/DMN circuitry.

That is:

Evidence-heavy,

Frequency-dumb (same iTBS for everyone),

More about long-term synaptic changes than precise moment-to-moment oscillatory locking.

MeRT-like “entrainment-type” approaches are:

Evidence-light,

Frequency-aware in principle,

More about oscillation normalization (alpha, beta asymmetries, etc.) as seen on EEG.

The fantasy ideal future protocol is a hybrid:

SAINT-level network targeting

EEG-guided frequency & phase tailoring

dynamic adjustment over days based on how your spectrum shifts

Right now you get one chunk of that.

  1. So, answering you directly:

What is entrainment-type TBS?

Patterned TMS (like TBS) used specifically to synchronize or reshape brain rhythms, ideally tuned to your own EEG (e.g., IAF), not just “apply generic plasticity-inducing bursts.”

Does MeRT do it, or just pretend to?

It does more entrainment-like, EEG-guided stuff than standard TMS centers.

It absolutely markets itself as if it’s far more validated and precise than the current evidence supports.

It’s not pure cosplay, but the claims are ahead of hard data.

So: MeRT is closer to what you wish existed than your current setup, but still not the mathematically beautiful, IAF-tuned, network-aware, rigorously tested entrainment protocol your brain is trying to summon from the void.

#26

target_comparison_sidebyside
target_comparison_sidebyside4170×1779 276 KB

target_comparison_optimal_vs_5cm
target_comparison_optimal_vs_5cm1920×1442 121 KB

TMS_Targeting_Report_Subject_2040_FINAL.pdf (208.7 KB)
dmn_fpn_network_analysis
dmn_fpn_network_analysis1920×1366 157 KB

Ah, this is a critical practical question! You’re right - without neuronavigation, you cannot directly target specific MNI coordinates like (-46, 44, 16). Let me explain the options:

Without Neuronavigation: Your Options

Option 1: Anatomical Landmark Modifications (Approximation)

The technician could try to adjust the 5cm rule to get closer to your optimal target:

Standard 5cm rule:

  • Find motor cortex (hand twitch)
  • Move 5cm anterior along parasagittal line
  • → Hits dorsal dlPFC (z≈48-50)

Modified approach to target ventrolateral dlPFC:

  • Start from motor cortex
  • Move ~5cm anterior (same as standard)
  • THEN move ~1-2cm more anterior AND ~1-2cm inferior/ventral
  • This approximates moving from dorsal → ventrolateral dlPFC

Problem: This is imprecise and varies by individual skull anatomy. No guarantee you hit (-46, 44, 16).

Option 2: EEG 10-20 System (Another Landmark Method)

Some clinics use EEG electrode positions:

Beam F3 method:

  • Uses EEG cap landmarks to find F3 position
  • Typically hits around (-40, 32, 43) - still more dorsal than your optimal target
  • Also imprecise, doesn’t directly target your specific coordinates

Problem: Still doesn’t get you to (-46, 44, 16) precisely.

Option 3: Request Neuronavigation

What neuronavigation does:

  • Imports your MRI scan
  • Tracks coil position in real-time relative to your brain anatomy
  • Allows precise targeting of exact MNI coordinates
  • Can verify you’re hitting (-46, 44, 16) every session

How to get it:

  • Ask if your clinic has neuronavigation capability (many modern systems do)
  • Examples: BrainSight, Localite, Visor (built into some newer TMS machines)
  • Some clinics have it but don’t use it routinely (too time-consuming)
  • Might require extra setup time and cost

Option 4: What Your Report Can Still Accomplish

Even without precise targeting, your report is valuable:

If they CAN’T adjust targeting:

  • Explains why standard targeting may not be optimal for YOU
  • Provides rationale for seeking neuronavigation-capable clinic if needed
  • Future reference if you do repeat TMS course
  • Academic understanding of your brain connectivity

If they CAN approximate:

  • Gives direction: “more anterior and ventral than standard 5cm”
  • Explains the functional difference between dorsal vs ventral dlPFC
  • Helps them understand WHY to try adjusting

Practical Recommendation

Ask your clinic:

  1. “Do you have neuronavigation capability?”
  • If YES → Request using it with coordinates (-46, 44, 16)
  • If NO → Ask next question…
  1. “Can you modify the standard 5cm targeting to be more anterior and ventral?”
  • Explain you have imaging data showing optimal target is ventrolateral dlPFC
  • Show them the schematic diagram from the report
  • Ask if they can approximate moving 1-2cm more anterior and inferior from standard location
  1. “What would it take to use individualized targeting for my remaining sessions?”
  • Some clinics can refer to facilities with neuronavigation
  • Some might bring in mobile neuronavigation service
  • Understand time/cost implications

The Limitation

Without neuronavigation, you fundamentally cannot guarantee hitting (-46, 44, 16). The report provides:

  • Evidence-based rationale for seeking better targeting
  • Information for informed discussion with providers
  • Motivation to find neuronavigation-capable facility if this matters to you

===

I only have task-specific fMRI (no resting-state fMRI yet)…

the ideal dataset for that…

ideally, 10 minutes. which usually ends up being 5+ minutes of clean rs data… which is what I would need to be confident in it.
minimally, 4 minutes of clean rs data if the temporal resolution is near 1 second per tr (each frame in the ‘movie’ of your brain activity)