Theta waves the 4–8 Hz rhythm of drowsiness, memory and meditation

What are theta waves?

Theta waves are the rhythms in a brain recording that repeat between 4 and 8 times a second. An EEG records tiny voltage changes, a few to about a hundred millionths of a volt, from electrodes on the scalp, usually placed by the international 10–20 system. The signal comes mainly from the synaptic currents of large groups of cortical neurons working in step, not from single cells.

That signal always contains many frequencies at once. Slow and fast rhythms run together, and the slower ones are larger: power falls steadily as frequency rises, with rhythmic peaks standing out above that slope. Delta, theta, alpha, beta and gamma are agreed slices of this one spectrum. Theta is the second-slowest slice, between delta below and alpha above.

So theta is a feature of a recording, not a mood or a mental mode. A person does not switch from 'alpha' into 'theta'; at any moment the EEG holds activity in every band, and what changes is how much there is in each and where on the head it shows. Even researchers say that much is still unknown about exactly how scalp EEG arises.

The interactive map above shows the five bands as moving traces, slowed four times, with relative heights, and a schematic EEG of six states from relaxed wakefulness to REM sleep with the share of each band; on this page it also shows a schematic hypnogram of a typical night. All of it is a drawing to explain the idea, not a real recording.

Theta wave frequency: 4 to 8 Hz, and why sources differ

The international clinical glossary (Kane and colleagues, 2017, for the International Federation of Clinical Neurophysiology) defines the theta band as 4 to under 8 Hz. That is the range used on this site. Band edges, however, are a convention. A review of 184 resting-EEG studies (Newson and Thiagarajan, 2018) found that the most common definition of each band was used by only 30–50 % of studies; for theta, the most common choice was 4–7.5 Hz, and definitions ranged from 2.5 to 8 Hz.

This matters whenever you read that a study found 'more theta'. Two papers can mean slightly different things by the word. The table compares the clinical definitions with what researchers actually use.

The German psychiatrist Hans Berger recorded the first human EEG in 1924 and published it in 1929, naming the first two rhythms he saw alpha and beta. Delta, theta and gamma were named later, as EEG research spread in the 1930s and 1940s.

Theta waves during sleep: drowsiness, N1 and REM

Theta is the band of falling asleep. When you lie relaxed with your eyes closed, the EEG at the back of the head is dominated by the alpha rhythm, 8–12 Hz in adults. As drowsiness sets in, alpha falls in frequency and amplitude, central theta of 5–8 Hz increases, the eyes make slow rolling movements, and short vertex sharp waves appear over the top of the head. A sleep lab scores this as stage N1 (Alcala-Zermeno and colleagues, 2025).

In stage N2, sleep spindles and K-complexes appear; in N3, deep sleep, large slow waves of 0.5–2 Hz take over. REM sleep brings the EEG back towards a waking look: low-amplitude, mixed-frequency waves mostly in the theta to alpha range, together with rapid eye movements and slack muscles. Up to 60 % of patients in a clinical review show sawtooth waves in REM, serrated runs of 2–6 Hz at the top of the head.

Sleep runs in cycles of about 90 minutes, four to five per night. Most deep sleep comes in the first half; REM periods grow longer towards morning (Steiger and Pawlowski, 2019). That is why theta-rich REM fills more of the second half of the night.

Theta also tracks sleepiness while you are still awake. A 1999 review by Klimesch notes that theta rises, and upper alpha falls, with sleep deprivation and when people drift towards sleep.

The shares come from a 2025 clinical review and change with age; older people have less deep sleep. The delta waves page covers deep sleep in detail.

Theta rhythm in the hippocampus: what rats showed

Much of the fame of theta comes from animal research. In 1969 C. H. Vanderwolf linked the rhythm of the rat hippocampus, a deep structure at the centre of research on memory and navigation, to voluntary movement. Later work sharpened the link: in their rats, hippocampal theta at 7–12 Hz went best with movements that change the animal's location (O'Keefe and Recce, 1993).

The same study described a striking timing code. Place cells in the hippocampus fire when the rat is in one particular spot. As the rat crosses that spot, the cell fires a little earlier in each theta cycle, a shift called phase precession. The phase of the theta wave at which the cell fires therefore carries information about where exactly the animal is.

Humans are different. Recordings from electrodes placed in the brains of neurosurgical patients show that 4–10 Hz hippocampal oscillations are rare in people; the functional equivalent of rat theta runs slower, at about 1–4 Hz (Jacobs, 2014). So 'theta' in rat studies and 'theta' on a human scalp EEG are not the same thing, and findings do not transfer directly from one to the other. Scalp EEG also cannot see the hippocampus directly: deep structures do not contribute to it in any direct way.

Frontal midline theta: theta waves, memory and concentration

In humans, the best-known waking theta is a rhythm over the midline of the forehead. A 2008 review (Mitchell and colleagues) concluded that frontal midline theta is correlated with working memory and sustained attention, and that it has also been studied in navigation, episodic memory, inward-directed attention and meditation.

The same review adds three cautions. Frontal midline theta is sporadic and of low incidence in the population, so many people simply do not show it. It increases with anti-anxiety drugs and in people with lower trait anxiety. And definitive evidence that it plays a functional role in thinking or emotion is still lacking.

This is very different from the waking theta that a neurologist worries about. In an awake adult, theta should not exceed about 30 % of the recording, and theta outside drowsiness and light sleep is considered abnormal on a clinical EEG, most often as part of a general disturbance of brain function (Alcala-Zermeno and colleagues, 2025). Brief, localised theta during a demanding task and widespread slowing at rest are two separate findings.

Theta waves and meditation: what EEG studies found

Meditation is where theta is most often mentioned, and here the research does show a pattern. A large review by Cahn and Polich (2006) described an overall slowing of the EEG with meditation, with theta and alpha activation related to how proficient the practitioner is; the authors also noted that effects are variable.

A later systematic review (Lomas and colleagues, 2015) gathered 56 studies with 1,715 participants, 1,358 of them healthy and 357 with psychiatric diagnoses. Mindfulness was most commonly associated with more alpha and theta power than eyes-closed rest, although such outcomes were not uniformly reported, and earlier studies gave a mix of increases, decreases and no difference. No consistent pattern was found for beta, delta or gamma.

Read the direction carefully. These studies show that meditation tends to go with more theta, especially in experienced practitioners. They do not show that producing theta by some other means makes a person meditate, and they do not make theta a 'meditation frequency'.

Is there a theta state? Theta music, beats and 7.83 Hz

'Theta state' is a popular phrase, not a term from clinical EEG. Because several rhythms run at once, a person is never purely 'in theta'; and since frontal theta is absent in many people and meditation results vary, theta is not something you can switch on like a light.

Music sold as 'theta music' rests on the same idea. The brain does lock onto music: its slow delta and theta activity, below 8 Hz, follows the note rate of a piece across ordinary tempi, and more strongly in musicians (Doelling and Poeppel, 2015). That tracking is set by the tempo, not by the label. No study has shown that a track labelled 'theta' moves the brain into the theta band.

Binaural beats at theta frequencies are the most researched sound in this area. A 2023 systematic review found 14 EEG studies of binaural beats: five found the brain followed the beat, eight did not, one was mixed (Ingendoh and colleagues). A 2019 meta-analysis of 22 studies found a medium effect on memory, attention, anxiety and pain (g = 0.45), which shows people respond to beats but not that their brain waves changed (Garcia-Argibay and colleagues). The page about binaural beats for meditation covers listening in detail and has the generator.

One more number often attached to theta is 7.83 Hz, the Schumann resonance. Lightning around the world makes the space between the ground and the ionosphere ring at about 7.8 Hz (Price, 2016). By the clinical definition that falls inside the theta band, just below alpha. The overlap is numerical; evidence that the resonance shapes human brain activity is limited to correlational studies, and the site's daily Schumann resonance page explains the physics.

Theta waves in babies and children

Children have much more theta than adults, and that is normal. The main resting rhythm at the back of the head first appears at 3–4 months of age and is slow. In a reference table quoted in a 2025 clinical review, its average frequency is 5.3 Hz up to one year, 6.8 Hz at two to three years, 7.9 Hz at four to five, 8.7 Hz at six to seven and 9.9 Hz in adults aged 16–50.

In other words, a baby's main waking rhythm runs at a speed that would count as theta in an adult; it speeds up through childhood and reaches the adult alpha range at school age. The change goes on later too: in 80 children aged 8–12, delta and theta decreased and alpha and beta increased with age (Clarke and colleagues, 2001).

Infants and children also show high-amplitude bursts of delta and theta at the moments of falling asleep and waking up, called hypnagogic hypersynchrony. It is a normal pattern at that age. This is why EEG reports are always read against age norms: the same amount of theta can be ordinary in a five-year-old and abnormal in a forty-year-old.

Theta/beta ratio and ADHD: can an EEG diagnose it?

The theta/beta ratio compares the power of slow theta with faster beta activity, usually at the electrode on the top of the head. For years it was proposed as a marker of ADHD, because some children with ADHD show relatively more theta.

On 15 July 2013 the US Food and Drug Administration classified the NEBA System, a device that uses the theta/beta ratio measured at electrode Cz in patients aged 6–17, combined with a clinician's evaluation, as an aid in diagnosing ADHD. The FDA stated that it is not to be used as a stand-alone test, only as confirmatory support after a completed clinical evaluation.

The same year, a meta-analysis (Arns, Conners and Kraemer, 2013) pooled 9 studies with 1,253 children with ADHD and 517 without. The pooled effect sizes, 0.75 and 0.62, were judged misleading overestimates because the studies differed so much; the gap had shrunk over the years because the ratio rose in the children without ADHD. The authors concluded that an excessive theta/beta ratio cannot be considered a reliable diagnostic measure of ADHD, though a subgroup of children does deviate.

In 2016 the American Academy of Neurology published a practice advisory (Gloss and colleagues). It said the theta/beta ratio and frontal beta power should not replace a standard clinical evaluation, warned of a risk of significant harm from misdiagnosis because of an unacceptably high false-positive rate, and advised against using the ratio to confirm a diagnosis outside research. ADHD is diagnosed by a clinician, not by a brain-wave reading.

Theta neurofeedback: what the controlled trials show

Neurofeedback is EEG biofeedback: a person sees or hears a signal made from their own brain waves, for example a game that runs when the theta/beta ratio falls, and tries to shift it. Theta/beta training for ADHD is its most studied form.

A 2016 meta-analysis by the European ADHD Guidelines Group (Cortese and colleagues) covered 13 randomised trials with 520 children and adolescents. Neurofeedback had a significant effect only on ratings by the people closest to the treatment, the least blinded (total symptoms, SMD 0.35). On probably blinded ratings, in trials with active or sham controls, and on laboratory tests of attention and inhibition, the effect was not significant.

The largest double-blind trial then settled the question for theta/beta training. The Neurofeedback Collaborative Group randomised 144 children aged 7–10 with moderate to severe ADHD, 142 analysed, to real neurofeedback or to a sham that looked identical but fed back prerecorded EEG. Both groups improved strongly (d = 1.5), but real neurofeedback was no better than sham at the end of treatment (d = 0.01) or at 13 months (2021). At 25 months there was still no difference, and the authors concluded that most of the improvement reflects nonspecific effects of the sessions rather than theta/beta training itself (2023).

Alpha-theta neurofeedback, sold for addiction, anxiety and creativity, traces back to a 1989 study in which alcoholics reported fewer relapses after training (Peniston and Kulkosky). That study had small groups and no sham-feedback control, and no sham-controlled trial or meta-analysis has confirmed it.