Skip to main content
Psychedex

Cranial electrotherapy stimulation (CES)

technology · Secular / clinical

Cranial electrotherapy stimulation (CES) is a neurotechnological practice that applies low-intensity pulsed microcurrent to the head via ear-clip or temple electrodes. It is the modern descendant of mid-twentieth-century Soviet electrosleep research, now FDA-cleared in the United States for anxiety and insomnia. It alters consciousness primarily through pulsed microcurrent — sub-milliamp to low-milliamp electrical stimulation proposed to modulate brainstem and limbic circuitry.

The practitioner attaches electrodes either to the earlobes (ear-clip devices) or to the temples (forehead-pad devices) and runs the device for a prescribed session. Current is delivered as a low-intensity pulsed waveform — micro- to milliamp range, with device-specific frequencies. Sessions are device-passive: the user sits or reclines while the device runs, typically for twenty to sixty minutes.

A typical session lasts twenty to sixty minutes. At sub-sensory intensities, most users feel nothing during stimulation; at perceptible intensities, the principal sensation is a mild tingling, pulsing, or warmth at the electrode contact. Some users report drowsiness, mild lightheadedness, or a quiet, settled state during and shortly after sessions. Boredom, restlessness, transient headache, and skin irritation at the contact points are the common unsought effects.

Adverse effects across published trials are mild and infrequent: skin irritation, transient headache, dizziness, and occasional contact burns from poorly maintained electrodes.[1] Demand cardiac pacemakers, pregnancy, active seizure disorder, and metal head implants are standard contraindications. The strongest controlled evidence supports anxiety reduction, with two meta-analyses reporting moderate-to-large effect sizes;[2][3] the definitive Alpha-Stim-D trial in primary major depression found no significant separation from sham.[4]

History & Tradition

Origins

The earliest documented application of low-intensity cranial electrical stimulation is attributed to Leduc and Rouxeau in France in 1902, who passed pulsed current through the skull as a therapeutic experiment. The technique remained a European curiosity for half a century. Beginning in 1953, Soviet clinicians systematically applied roughly 100 pulses per second at 1–4 mA to induce relaxation and sleep, naming the procedure elektrosen ("electrosleep").[1] Soviet electrosleep research generated a substantial Russian-language clinical literature through the 1950s–1970s, only partially absorbed by Western medicine.

American adoption began in the 1960s, when entrepreneurs observed Japanese use of an electrosleep device (the Somniatron) for railroad-operator fatigue management and imported units to Texas. Psychiatrists at the University of Texas Medical School in San Antonio ran early American trials in anxious and insomniac patients with thrice-weekly 30-minute sessions.

Transmission and commercial development

Saul Liss, an engineer in Paterson, New Jersey, built the Liss Pain Suppressor in the 1970s using a composite waveform — a 15 kHz carrier, a 15 Hz active wave, and a 500 Hz modifier — and obtained FDA clearance for depression and later for pain. After Liss's death in 2006, the device rights passed to Fisher-Wallace Laboratories of New York, which markets the Fisher Wallace Stimulator (5, 500, and 15,000 Hz; 1–4 mA; temple-pad electrodes) as one of the two principal US CES platforms today.

The second major US platform, the Alpha-Stim, manufactured by Electromedical Products International of Mineral Wells, Texas, delivers a substantially lower current (10–500 μA at 0.5 Hz) via ear-clip electrodes. The Alpha-Stim has been the subject of more controlled trials than any other CES device, including the definitive Alpha-Stim-D depression trial.[4] The Alpha-Stim and Fisher Wallace waveforms are not interchangeable, and trial results from one device do not transfer to the other without independent evidence.

When the FDA first cleared the modality in the United States, it renamed it from "electrosleep" to "cranial electrotherapy stimulation," shifting the nomenclature from intended outcome to nominal mechanism; "electrosleep" remains the common term outside North America.

Contemporary context

In December 2019, the FDA finalized a reclassification: CES devices for anxiety and insomnia were placed in Class II under special controls (510(k) pathway), while CES devices marketed for depression were assigned to Class III, requiring premarket approval (PMA).[5] The agency explicitly concluded that the evidence for depression was insufficient to support 510(k) clearance, and no CES device currently holds PMA for depression in the United States.

Interest in CES expanded significantly through the 2010s as the US Department of Defense and Department of Veterans Affairs authorized CES prescriptions for active-duty service members and veterans with anxiety, PTSD, insomnia, depression, and headache — a population for whom pharmacological side-effect burden is a recognized clinical concern. Major commercial insurers (BCBS, Aetna) continue to classify CES as investigational for most psychiatric indications and exclude it from standard coverage; the VA remains the principal channel of funded access.

The Practice

Technique

The practice is a device-administered protocol with no skill component. The user attaches electrodes, sets the device parameters, and sits through a fixed-duration session. The variation in practice is between devices and electrode placements, not between practitioners.

The traditional posture is seated upright in a chair or reclined comfortably on a couch or bed. The session is sensory-permissive — the user may read, watch a screen, or close their eyes during stimulation. No special environment is required beyond reasonable quiet. Loose, dry skin at the contact points and clean electrode pads are the only physical prerequisites.

The core method is the delivery of low-intensity pulsed electrical current to the cranium for a fixed period. In ear-clip designs (Alpha-Stim and similar), conductive felt pads moistened with conducting solution are clipped to both earlobes; in temple-pad designs (Fisher Wallace Stimulator and similar), saline-moistened sponge electrodes are positioned bilaterally on the temples and held by a headband. The device generates a pulsed waveform — micro- to milliamp peak current, device-specific frequencies — for the prescribed session length. Intensity is titrated by the user: increased until a faint tingling or pulsing sensation is perceived, then often reduced slightly to a sub-sensory level.[6] The most common departure from the method is using the device at the wrong intensity — pushing the current high enough to produce discomfort under the assumption that more sensation means more effect, when controlled studies have used sub- or just-perceptible intensities.

There is no progressive structure to CES in the meditative sense. Clinical protocols converge on daily 20–60-minute sessions for two to eight weeks; the Alpha-Stim-D depression trial used daily 20-minute sessions for eight weeks,[4] and the anxiety meta-analyses pool trials of similar duration.[2] Some clinicians report that responders typically show change within two to four weeks and consider non-response at six weeks a basis for re-evaluation. Beyond the prescribed course, "deepening" the practice means adhering to the daily schedule, not advancing through stages.

Variants

The two principal US devices differ substantially in waveform parameters. The Alpha-Stim AID delivers 10–500 μA at 0.5 Hz via ear-clip electrodes and is the device used in most modern controlled trials, including both the positive anxiety meta-analyses and the negative Alpha-Stim-D depression trial.[3][4] The Fisher Wallace Stimulator delivers a composite waveform at 5, 500, and 15,000 Hz at 1–4 mA via bilateral temple electrodes — current orders of magnitude higher than the Alpha-Stim. The CES Ultra, NET Device, and several other cleared devices occupy intermediate parameter ranges. Because device waveforms differ in current intensity, frequency content, and electrode placement, trial results from one device do not generalize to the others without independent replication.

Setting & Preparation

A CES session requires the device, the manufacturer-specified electrodes (ear clips or temple pads), conducting solution or saline-moistened sponges, and a power source. No special environment is required beyond reasonable quiet and a comfortable seat or recliner. The practice is fully solo — no facilitator, group, or instructor is required during a session — though in the United States a prescription from a licensed clinician is required to obtain a Class II CES device.[7]

Set up a clean, quiet space and a comfortable chair or recliner where you can sit through a twenty-to-sixty-minute session without being disturbed. Inspect the electrode pads before each session: dried, cracked, or worn pads are the dominant cause of contact burns and should be replaced per the manufacturer's schedule.[1] Moisten the pads with the manufacturer-specified conducting solution or saline; dry contact is both ineffective and the principal route to skin irritation. Confirm the device is at the correct intensity setting from the previous session before starting — many devices retain settings between uses.

Guidelines

  • Obtain CES through a licensed clinician. In the United States, CES devices are prescription-only medical devices; the clinical evaluation screens for contraindications and matches device parameters to the indication.
  • Do not use CES if you have a demand cardiac pacemaker, are pregnant, have an active seizure disorder, or have metal implants in or near the head.
  • Start at the lowest device intensity and titrate upward only to the threshold of faint perception. Sub-sensory and just-perceptible intensities are the levels used in controlled trials; higher current does not mean greater benefit.
  • Use fresh, properly moistened electrode pads every session. Replace pads per the manufacturer's schedule. Electrode-site burns are the dominant serious adverse event and are almost entirely a pad-maintenance failure mode.
  • Hold the device's recommended protocol — typically daily 20-minute sessions for several weeks — rather than improvising frequency or duration.
  • Re-evaluate with your prescribing clinician at six weeks. Non-response by that point is a reasonable basis for reconsidering the intervention.
  • Do not combine CES with opioid analgesics outside a clinical context. Perioperative data indicate opioid potentiation; outpatient interaction effects are uncharacterised.

Mechanism

CES is classified as a neurotechnological practice. Its proposed mechanisms are multiple and contested; the 2018 VA Evidence-based Synthesis Program review concluded that the physiological, neurochemical, and metabolic mechanisms underlying CES effects are not currently known.[1]

Current distribution and subcortical reach

Datta and colleagues constructed a high-resolution finite-element model of CES current propagation through head anatomy and found that current penetrates to subcortical structures, with the strongest modeled neuromodulatory effects in the temporal lobes, medulla oblongata, midbrain, thalamus, hypothalamus, pons, and insula rather than primarily in cortex.[8] This computational result is consistent with the hypothesis — though does not demonstrate — that CES effects are mediated through brainstem and limbic circuits, not through direct cortical stimulation.

Neuroimaging effects

Feusner and colleagues scanned eleven healthy controls during active Alpha-Stim CES at sub-sensory threshold and observed statistically significant resting-state deactivation across midline cortical and parietal regions: bilateral pre- and postcentral gyri, right posterior cingulate cortex, right lateral occipital cortex, and bilateral precuneus.[6] Both 0.5-Hz and 100-Hz stimulation produced the pattern. Default mode network connectivity was altered. The deactivation signature resembles cortical patterns associated with anxiolytic medications and sedation. The study used only healthy controls and a single-session design, and generalization to clinical populations remains untested.

EEG and oscillatory effects

Studies recording EEG concurrently with CES have observed a downward shift in mean alpha frequency and modulation of alpha and beta band power, interpreted as reflecting reduced cortical arousal.[9] CES does not produce reliable entrainment of brain rhythms to the stimulation frequency in the manner sometimes claimed by binaural beat proponents; the alpha-band changes are interpreted as a global arousal-reduction response rather than frequency following. Whether these EEG changes mediate clinical effects has not been demonstrated.

Proposed neurochemical mechanisms

Several neurochemical hypotheses have been advanced. Proponents claim that CES applied at the maxillo-occipital junction routes current to brainstem structures and increases secretion of serotonin, beta-endorphins, and norepinephrine while reducing cortisol. The 2021 critical review notes that most subsequent studies have not replicated these neurochemical effects, and an earlier positive report on beta-endorphin and cortisol changes remains unreplicated.[9] A vagal-trigeminal afferent hypothesis is also proposed: ear-clip electrode placement is anatomically adjacent to the auricular branch of the vagus nerve and trigeminal nerve terminals, and stimulation of these afferents is a plausible pathway by analogy with auricular acupuncture and transcutaneous vagal nerve stimulation. The autonomic-modulation hypothesis — that CES increases parasympathetic relative to sympathetic drive — is consistent with some heart-rate-variability findings but has not been consistently replicated.

The mechanism by which any of these proposed pathways might produce the moderate anxiolytic effect seen in controlled trials remains unverified. The proposed mechanisms are best read as competing hypotheses awaiting adequately powered replication.

Subjective Effects

CES is a quiet practice. Most reported effects are subtle, accumulating over a course of daily sessions rather than appearing dramatically during a single session. The principal sought effect across the controlled literature is reduction in anxiety symptoms, with moderate-to-large effect sizes reported in two meta-analyses.[2][3] Improvements in sleep latency and subjective sleep quality are also reported in trials of CES for primary insomnia and for insomnia comorbid with anxiety. The second 2023 meta-analysis found significant effects on co-occurring depression and insomnia in primary-anxiety populations.[3]

Within-session sensations are typically minimal. At sub-sensory intensities — the level used in much of the controlled-trial literature — users feel nothing. At perceptible intensities, the principal sensation is a faint tingling, pulsing, or warmth at the electrode contact points. Some users report a quiet, settled feeling during and shortly after sessions; others report mild drowsiness or transient lightheadedness. Boredom and restlessness during the twenty-to-sixty-minute device-passive session are common, particularly at the introductory tier. Headache and skin irritation at the electrode sites are the most commonly reported adverse sensations and resolve when stimulation is stopped or current is reduced.[1]

▸Effect Grid
Intensity Level:
introductory

Ear-clip microcurrent session for anxiety/sleep.

Body(8)intensity
Restlessnesslikely
Tingling / electric sensationslikely
Spontaneous body sensationslikely
Vibrations / buzzinglikely
Sedationpossible
Physical fatiguepossible
Headachepossible
Dizzinessrare

Safety

Adverse Effects

Frequency and character

Adverse effects in published CES trials are consistently mild and infrequent, with most estimates placing event rates below 1% of treatment sessions.[1] The 2018 VA Evidence-based Synthesis Program review concluded that CES is "probably safe" based on available trial data, while explicitly noting likely underreporting bias given small sample sizes and inadequate safety reporting in many included studies.[10]

The most commonly reported events are skin irritation, stinging, or burning at the electrode contact sites (earlobes or temples); transient tingling during stimulation; mild transient headache (fewer than 1 in 250 patients in some estimates); and transient dizziness. These effects are self-limiting and resolve when stimulation is stopped or the current reduced.

Serious events

Review of the FDA adverse-event database between 2012 and 2019 identified three reported serious-level event categories: electrode-site burns on the earlobes, new onset of severe tinnitus, and severe gastrointestinal distress coincident with insomnia. The electrode-site burns are attributable to reused or dried-out electrode pads — a device-handling failure mode rather than an inherent stimulation risk — and are largely preventable with proper electrode maintenance. No serious adverse events were observed in any RCT included in the 2018 VA systematic review.[1]

Contraindications

Published contraindications for CES are drawn from device labelling and the FDA classification regulation[7]:

  • Demand cardiac pacemakers — risk of electrical interference with pacemaker function.
  • Pregnancy — insufficient safety data; no trials in pregnant populations.
  • Active seizure disorder — theoretical proconvulsant risk; no documented cases in the published literature, but the contraindication is retained on caution.
  • Metal implants in or near the head — perturbation of current distribution.
  • Application over broken, infected, or swollen skin at the electrode contact sites.

In the United States, CES devices are FDA-regulated medical devices requiring a prescription from a licensed healthcare practitioner.[7] A clinician evaluation prior to initiation is the standard route through which contraindications are screened.

Substance Interactions

Documented pharmacodynamic interactions between CES and psychoactive substances are sparse. No peer-reviewed studies specifically designed to characterise CES–substance interactions in outpatient or recreational contexts have been published; the existing literature is largely confined to perioperative and addiction-treatment contexts.

The best-characterised interaction is with opioid analgesics. In experimental and perioperative settings, CES has been observed to potentiate opioid analgesic effects, measurably reducing anesthetic requirements when co-administered. The proposed mechanism is endogenous opioid peptide release stimulated by CES. The implication for outpatient practice — whether CES combined with therapeutic or recreational opioid use produces additive central-nervous-system depression — has not been studied. Early controlled work by Patterson examined CES in opiate-dependent patients undergoing detoxification with mixed results: some CES-treated subjects experienced more severe early withdrawal than methadone-treated controls; others showed comparable or superior outcomes.[11] A more recent randomised sham-controlled trial of the NET Device (a CES variant) for opioid discontinuation symptom severity reported preliminary positive findings.[12]

No published case reports or controlled studies have documented interactions between CES and alcohol, stimulants, cannabis, or serotonergic agents. Given the proposed parasympathomimetic and cortical-deactivation effects of CES, the theoretical direction of an interaction with central-nervous-system depressants would be additive sedation; this remains unverified and represents a genuine evidence gap rather than a documented safety signal.

Habituation

CES carries no known habituation or compulsive-use risk. The practice is device-mediated, prescription-gated in the United States, and produces no acute reinforcing experience comparable to a substance high. There are no published case reports of compulsive CES use or device dependency. Tolerance — in the sense of diminishing effect with continued daily use — has not been characterised in controlled trials, and protocols of two to eight weeks do not provide the duration to detect it. When users stop CES at the end of a treatment course, no withdrawal syndrome has been documented.

If Distress Arises

Stop the session if you experience pain, burning, or significant discomfort at the electrode sites — these are signals of either current too high or a dried or worn pad, and they will not abate by pushing through. Reduce the current first; if discomfort persists, end the session and inspect the pads before the next use. Lightheadedness or strong dizziness during a session is uncommon but is a reason to stop and rest. If you experience a seizure, severe headache that does not resolve, or any sustained cardiovascular symptom, stop the device and seek medical attention. Because CES is user-controlled, ending the session at any time is straightforward and carries no rebound effect.

Recovery & Support

CES does not produce a compulsive-use pattern requiring exit support. The relevant recovery context is clinical: if you are using CES for anxiety, depression, insomnia, or PTSD and your symptoms worsen or fail to improve, contact your prescribing clinician rather than self-titrating the device. For acute mental health crises during a course of CES treatment — suicidal ideation, severe panic, psychotic symptoms — standard crisis resources apply (988 Suicide and Crisis Lifeline in the United States; equivalent national lines elsewhere). The device is a peripheral instrument in any treatment plan; do not let its presence substitute for ongoing clinical contact when symptoms shift.

Recovery resources, crisis lines, and support pathways → View all

Progression

CES has no progression model in the developmental sense. The practice is a device-administered protocol, not a skill, and there is no published framework analogous to jhana progress maps, kundalini stages, or yoga forms.[1] What changes over a treatment course is symptom response, not depth of practice. Clinical protocols in published trials converge on daily 20-minute sessions for two to eight weeks; the Alpha-Stim-D depression trial used eight weeks of daily sessions as its treatment arm,[4] and the controlled anxiety literature pools trials of similar duration. Clinicians who prescribe CES commonly report that responders show change within two to four weeks; non-response at six weeks is generally treated as a basis for re-evaluation. There is no "advanced" CES practice — only adherence to a prescribed course.

Evidence Summary

The overall evidence base for CES is moderate in volume but heterogeneous in quality. Trials are limited by small sample sizes, varied devices and protocols, difficulty maintaining blinding (current sensation is detectable above sub-sensory thresholds), and short follow-up. The 2018 VA Evidence-based Synthesis Program review screened 1,924 citations and identified 28 eligible publications;[1] the post-2018 literature has added several meaningful trials and meta-analyses but has not transformed the evidence picture for most indications.

Anxiety

Anxiety is the best-supported indication. A 2022 meta-analysis of 11 RCTs (n=794) reported a moderate effect size of Hedge's g = −0.625 in favour of active CES versus sham, with good tolerability.[2] A 2023 meta-analysis of 8 RCTs (n=337) reported a larger pooled effect (ES = −0.96, p < 0.00001), with a subgroup analysis of four Alpha-Stim trials (n=230) showing a comparable effect (ES = −0.88).[3] The two meta-analyses used different inclusion criteria and partially overlapping trial sets; their estimates are not directly comparable but converge on a statistically significant, clinically moderate signal.

Depression

The depression picture is mixed and dominated by the negative Alpha-Stim-D trial. The trial — a multicentre, double-blind, parallel-group RCT in English primary care — found HDRS-17 changes of −5.9 points (active) versus −6.5 points (sham) at 16 weeks (p = 0.46) with no clinically meaningful separation.[4] The pre-registered protocol was published in 2022.[13] A 2025 meta-analysis pooling 16 studies (2006–2023) reported a small but statistically significant pooled effect across primary and secondary depression populations, with the negative Alpha-Stim-D trial contributing to heterogeneity.[14] The FDA concluded in 2019 that evidence for CES in depression was insufficient to support 510(k) clearance.[5]

Insomnia, PTSD, and other indications

A meta-analysis by Price and colleagues pooling RCTs and non-randomised studies of CES for insomnia reported beneficial effects on subjective sleep quality and sleep latency, consistent with the existing FDA clearance for insomnia.[15] PTSD evidence is limited to a 2022 open-label feasibility pilot in nine veterans, which reported an excellent safety profile and preliminary signals across PTSD severity, depression, insomnia, and pain — a hypothesis-generating result, not an efficacy result.[16] A pilot RCT in bipolar II depression reported a favorable safety profile and preliminary mood signal but was not powered for efficacy.[17] Evidence for CES in chronic pain, headache, and fibromyalgia was rated insufficient by the 2018 VA review.[1]

Limitations

Several methodological and structural limitations qualify the CES evidence base.

Blinding is difficult. CES at perceptible intensities produces a detectable tingling sensation at the electrodes; sham conditions delivering no current or sub-threshold current may be discriminable to participants, which can inflate effect estimates. Sham fidelity is reported inconsistently across the trial literature.[9]

Device heterogeneity confounds aggregation. The Alpha-Stim (μA, 0.5 Hz, ear-clip) and Fisher Wallace Stimulator (mA, 5/500/15,000 Hz, temples) differ by orders of magnitude in current intensity, contain different frequency content, and apply current at different anatomical sites. Pooling trials of different devices into a single "CES" effect estimate likely obscures device-specific responses. The 2023 anxiety meta-analysis's Alpha-Stim-only subgroup analysis is one attempt to address this, but most published meta-analyses aggregate across devices.[3]

Sample sizes are modest. The Alpha-Stim-D trial is the largest rigorously powered CES RCT to date and is negative for depression; most other trials in the literature are under-powered to detect clinically meaningful effects against well-managed sham conditions.[4]

Adverse-event reporting is inadequate. The 2018 VA review explicitly noted that the conclusion of "probably safe" rests on a trial literature with likely underreporting bias.[1]

Mechanism remains unestablished. None of the proposed mechanisms — current penetration to brainstem structures, alpha-band oscillatory modulation, endogenous opioid or monoamine release, autonomic shift — has been confirmed in adequately powered replicated studies.[9] The mechanism cannot currently be stated as established fact.

references

  1. [1] 
    ^abcdefghijkShekelle, P.G. and Cook, I.A. and Miake-Lye, I.M. and Mak, S. and Booth, M.S. and Shanman, R. and Beroes, J.M. (2018) Benefits and Harms of Cranial Electrical Stimulation for Chronic Painful Conditions, Depression, Anxiety, and Insomnia: A Systematic Review — Annals of Internal Medicine doi:10.7326/m17-1970
  2. [2] 
    ^abcde[Authors: see doi or PMC9218324 — first author not confirmed in search] (2022) Efficacy and Tolerability of Cranial Electrotherapy Stimulation in the Treatment of Anxiety: A Systemic Review and Meta-Analysis — Frontiers in Psychiatry doi:10.3389/fpsyt.2022.899040
  3. [3] 
    ^abcdef[Authors: see doi or PMC10115990 — first author not confirmed in search] (2023) Efficacy of electrical cranial stimulation for treatment of psychiatric symptoms in patients with anxiety: A systematic review and meta-analysis — Frontiers in Psychiatry doi:10.3389/fpsyt.2023.1157473
  4. [4] 
    ^abcdefghPatel, S. and Boutry, C. and Patel, P. and et al. (2023) Clinical effectiveness of active Alpha-Stim AID versus sham Alpha-Stim AID in major depression in primary care in England (Alpha-Stim-D): a multicentre, parallel group, double-blind, randomised controlled trial — The Lancet Psychiatry doi:10.1016/s2215-0366(23)00007-x
  5. [5] 
    ^abU.S. Food and Drug Administration (2019) Neurological Devices; Reclassification of Cranial Electrotherapy Stimulator Devices
  6. [6] 
    ^abFeusner, J.D. and Madsen, S. and Moody, T.D. and Bohon, C. and Hembacher, E. and Bookheimer, S.Y. and Bystritsky, A. (2012) Effects of cranial electrotherapy stimulation on resting state brain activity — Brain and Behavior doi:10.1002/brb3.45