Garcia-Argibay et al. (2019): Binaural Beats for Cognition, Anxiety and Pain – An In-Depth Meta-Analysis

Garcia-Argibay, Santed and Reales published an influential meta-analysis on binaural auditory beats in 2019. By statistically combining 35 effect sizes from 22 studies, the researchers examined whether binaural beats influenced cognition, anxiety and pain perception. They reported an overall statistically significant medium effect of g = 0.45.

This finding is encouraging, but its correct interpretation requires more than repeating one number. The meta-analysis combined different populations, frequencies, exposure schedules and outcomes. It therefore provides evidence for an average research signal, not a guarantee that every recording, beat frequency or listener will produce the same result.

The central lesson is not simply that binaural beats “work.” It is that their measured effects appear to depend on the selected frequency, exposure duration and timing in relation to the task.

The study at a glance

What exactly is a binaural beat?

A binaural beat is an auditory percept created when two pure tones with slightly different frequencies are presented separately to the left and right ear. If one ear receives 200 Hz and the other 210 Hz, the listener may perceive a rhythmic fluctuation corresponding to the 10 Hz difference. That 10 Hz rhythm is not physically present as an ordinary amplitude pulse in either channel; it arises through binaural auditory processing.

Because each ear must receive its own tone, stereo headphones are normally required. This distinguishes binaural beats from monaural beats and isochronic tones, whose physical amplitude modulation can be present in the acoustic signal itself and can therefore remain audible through loudspeakers.

The meta-analysis focused specifically on binaural auditory beats. Its conclusions should therefore not automatically be transferred to flashing light, monaural beats, isochronic stimulation or complete audiovisual mind-machine programmes. These may share a rhythmic principle but are technically different interventions.

Why a meta-analysis matters

Individual studies often include relatively small samples and can produce apparently conflicting findings. A meta-analysis converts results into a common statistical metric and combines them. This can estimate whether a pattern remains visible across multiple experiments and can investigate variables that might explain differences between studies.

However, a meta-analysis cannot repair every weakness in the source literature. If included studies are small, use weak blinding, differ greatly in design or selectively report favourable outcomes, the pooled result inherits part of that uncertainty. Statistical synthesis is powerful, but it is not a substitute for high-quality primary research.

What did the researchers investigate?

The authors examined four broad outcome areas: memory, attention, anxiety and analgesia. Cognition in the title mainly refers to memory and attention rather than every aspect of human thinking. Analgesia refers to reduced pain or reduced pain perception, not necessarily treatment of the underlying cause of pain.

• Memory: whether performance on memory tasks changed during or after binaural-beat exposure.
• Attention: whether listeners showed changes in concentration, vigilance or attentional task performance.
• Anxiety: whether self-reported or measured anxiety decreased in situations such as procedures or laboratory tasks.
• Pain perception: whether pain ratings or analgesic-related outcomes changed.

Understanding the result: Hedges’ g = 0.45

Hedges’ g is a standardised effect size. It expresses the average difference between conditions in units of standard deviation and includes a correction that is useful when studies have small samples. A positive value in this meta-analysis indicated an outcome in the intended direction.

Rules of thumb often label approximately 0.2 as small, 0.5 as medium and 0.8 as large. These boundaries are not laws of nature. An effect of 0.45 is best described as modest to medium: large enough to be scientifically interesting, but not so large that outcomes become predictable for each individual.

The number also represents an average across different outcomes. It does not mean that memory, attention, anxiety and pain each improved by exactly 45 percent. A standardised effect size is not a percentage, and translating it directly into a personal probability of success would be misleading.

Finding 1: exposure timing appeared to matter

The meta-regression suggested that exposure before a task, or both before and during a task, produced better outcomes than exposure only during the task. This is one of the most practically relevant findings in the paper.

One cautious interpretation is that the auditory context needs time to influence arousal, expectation or attentional state before performance is measured. Another possibility is that listening only during a demanding task introduces distraction. The meta-analysis identifies a statistical relationship, but it does not by itself prove which psychological or neurophysiological mechanism produced it.

Finding 2: longer exposure was associated with stronger effects

Time under exposure contributed significantly to the statistical model. The authors therefore advised longer exposure periods to maximise effectiveness. This does not establish one universal minimum duration, and it certainly does not mean that listening indefinitely produces ever-increasing benefits.

Duration may interact with frequency, task, participant characteristics and habituation. A sensible application uses a defined preparation period and monitors comfort and performance instead of assuming that more is always better.

Finding 3: masking with white or pink noise was not necessary

Some binaural-beat recordings place the tones beneath white noise, pink noise, music or ambient sound. The meta-regression indicated similar effectiveness for masked and unmasked beats. In the included evidence, masking did not appear necessary for the effect.

This does not make the carrier sound irrelevant. Music and noise may influence comfort, mood, expectancy and willingness to continue. The finding means that masking was not statistically required for effectiveness in this analysis, not that all sound design is psychologically identical.

Finding 4: frequency selection influenced direction and magnitude

The authors concluded that both the direction and magnitude of effects depended partly on the frequency used. This supports the idea that a protocol should be selected in relation to its intended outcome rather than treating every binaural beat as interchangeable.

It does not justify a simplistic frequency dictionary in which one exact hertz value guarantees one mental state. Studies differ in carrier tones, difference frequencies, instructions, participants and measurement methods. Frequency is one protocol variable among several.

What the meta-analysis suggests about cognition

The pooled evidence supported an average influence on cognitive outcomes, specifically memory and attention. This is relevant to studying, focused work and performance preparation. However, improved performance under selected experimental conditions is not the same as permanent cognitive enhancement.

A person can also respond differently depending on baseline arousal. Someone who is tired may benefit from a more activating context, while an already overstimulated person may perform worse. Individual state, preference and task demands remain important even when the average effect is positive.

What the meta-analysis suggests about anxiety

The anxiety findings are compatible with binaural beats being used as a non-pharmacological supportive method before stressful situations. Listening creates a structured pause and may interact with breathing, attention, expectation and the acoustic environment.

The evidence does not establish binaural beats as a stand-alone treatment for anxiety disorders, panic disorder, trauma or obsessive-compulsive disorder. Clinical symptoms require appropriate assessment, and supportive audio should not delay professional care.

What the meta-analysis suggests about pain

Pain perception involves sensory input, attention, expectation, emotion and context. Binaural beats may potentially alter the attentional or affective component of the experience. The included literature contained analgesia-related outcomes, including perioperative contexts.

Reduced pain perception is not evidence that tissue damage, inflammation or disease has resolved. Binaural beats must not replace medical evaluation, prescribed analgesia or treatment of an underlying condition.

Why the study was influential

Earlier reviews, including Huang and Charyton’s 2008 paper, organised a broad brainwave-entrainment literature. Garcia-Argibay and colleagues narrowed the question to binaural beats and added quantitative synthesis. That shift from a narrative overview to pooled effect sizes made the paper especially important.

The study also moved discussion beyond the question of whether an effect exists. Its moderator analyses asked how protocol design might matter: masking, timing, exposure duration and frequency. These are the kinds of variables that professional session design should document.

Important limitations

Twenty-two studies remain a modest evidence base

A statistically significant meta-analysis can still be based on a relatively small literature. Subgroup and moderator findings become less stable when only a limited number of studies contribute to each comparison.

The outcomes and contexts differed

Memory, attention, anxiety and pain are not one outcome. Laboratory cognitive tasks, dental anxiety and perioperative analgesia involve different participants, settings and mechanisms. An overall effect summarises them statistically but does not erase their clinical differences.

Blinding can be difficult

Participants may hear differences between recordings or develop expectations about brainwave technology. Control conditions can also differ in music, noise or sensory character. Expectancy is a genuine part of the listening experience and can influence subjective outcomes.

An average effect does not identify responders

Some listeners may improve, some may notice nothing and some may find the sound unpleasant or distracting. Group averages cannot predict an individual’s response without further assessment.

Psychological effects do not prove EEG entrainment

The paper evaluated behavioural and subjective outcomes. Even if those outcomes change, this does not automatically demonstrate that cortical oscillations synchronised to the beat frequency. Binaural-beat perception, arousal, attention and expectation could contribute through multiple pathways.

Publication date and later evidence matter

The article was published online in 2018 and in print in 2019. It cannot include newer studies. Later reviews have continued to report promising average outcomes, while also emphasising heterogeneity, inconsistent electrophysiological findings and the need for better-powered, preregistered trials.

How later research refined the picture

Basu and Banerjee’s 2023 meta-analysis on memory and attention reported a small-to-medium overall effect but noted conflicting findings for theta- and beta-frequency stimulation. Ingendoh and colleagues reviewed EEG entrainment evidence in 2023 and found a mixed pattern: some studies supported entrainment, while many were contradictory or inconsistent.

These later publications do not simply cancel the 2019 result. They help separate two questions: whether binaural-beat exposure can influence psychological performance or experience, and whether a consistent EEG frequency-following mechanism explains that influence. The evidence for the first question can be positive while the second remains unsettled.

What this study supports

• Binaural beats are a legitimate subject of quantitative psychological research.
• Across 22 studies and 35 effect sizes, the authors found a significant average effect.
• Memory, attention, anxiety and pain perception were all represented in the evidence base.
• Exposure timing and duration may be meaningful protocol variables.
• Masking with white or pink noise did not appear necessary for effectiveness.
• Frequency selection should be related to the intended application.
• Binaural beats require no complex prior training.

What this study does not prove

• It does not prove that every binaural-beat recording is effective.
• It does not show that every listener responds in the same way.
• It does not mean that g = 0.45 equals a 45 percent improvement.
• It does not demonstrate permanent intelligence or memory enhancement.
• It does not establish one exact frequency as universally optimal.
• It does not prove that psychological effects are caused by EEG synchronisation.
• It does not validate medical treatment claims for anxiety or pain disorders.
• It does not directly evaluate monaural beats, isochronic tones or light stimulation.

Responsible practical application

A research-informed protocol should define its objective, frequency, carrier sound, duration, timing, volume and listening equipment. It should also define what outcome will be observed. For focus, this might be task accuracy or distraction. For relaxation, it might be subjective tension before and after the session.

• Use stereo headphones with correct left-right channel separation.
• Begin at a comfortable, conservative volume.
• Allow a preparation period before the target task.
• Choose the protocol in relation to the intended state.
• Avoid changing several variables at once when evaluating response.
• Track both desired effects and unwanted distraction or discomfort.
• Treat individual response as data rather than as success or failure.

Explore practical designs in the NeuroSync Pro® categories Focus & Concentration, Relaxation, Meditation and Recovery & Well-Being.

Safety and medical boundaries

Audio-only binaural beats do not involve flashing light, but ordinary listening safety still applies. Keep volume moderate, protect hearing and never use attention-altering audio while driving, cycling, operating machinery or performing safety-critical work.

NeuroSync Pro® is not a medical device. Brainwave entrainment is not intended to diagnose, treat, cure or prevent anxiety disorders, chronic pain, neurological disease or cognitive impairment. Seek qualified care for persistent anxiety, significant pain, sudden headache, hearing symptoms or cognitive decline.

Frequently asked questions

Was this a single binaural-beat experiment?

No. It was a meta-analysis combining 35 effect sizes from 22 earlier studies.

What does g = 0.45 mean?

It is a standardised average effect commonly interpreted as modest to medium. It is not a percentage improvement and does not predict an individual result.

Did binaural beats improve intelligence?

The paper examined memory and attention outcomes, not permanent general intelligence. Short-term task effects should not be described as an increase in IQ.

Are longer sessions always better?

Longer exposure was associated with stronger effects in the meta-regression, but the paper did not establish unlimited benefit or one universal duration. Comfort, task and individual response still matter.

Must binaural beats be hidden under music or noise?

No. The analysis found similar effectiveness for masked and unmasked beats. Masking can still improve comfort or listening preference.

Can speakers be used?

True binaural presentation normally requires stereo headphones so that each ear receives a different tone. Loudspeakers allow acoustic mixing and are less suitable for controlled binaural delivery.

Does this prove brainwaves follow the beat?

No. The meta-analysis evaluated cognition, anxiety and pain outcomes. Evidence for consistent EEG entrainment is a related but separate scientific question.

Conclusion: meaningful evidence with important boundaries

Garcia-Argibay and colleagues made an important contribution by showing that the binaural-beat literature available to them produced a statistically significant average effect across cognition, anxiety and pain perception. The result of g = 0.45 is substantial enough to take seriously and modest enough to interpret carefully.

The most useful message for professional practice is that protocol design matters. Frequency, preparation time, duration and the moment of exposure may influence outcomes. The paper therefore supports structured experimentation and transparent monitoring more strongly than one-size-fits-all claims.

For NeuroSync Pro®, this means combining accessible technology with responsible expectations: define the goal, select an appropriate session, observe individual response and keep every claim proportional to the evidence. Learn more on the NeuroSync Pro homepage.

Primary publication and further reading

• Garcia-Argibay et al. (2019): PubMed record of the meta-analysis
• Garcia-Argibay et al. (2019): publication at Springer
• Huang & Charyton (2008): early review of psychological brainwave-entrainment outcomes
• Basu & Banerjee (2023): meta-analysis of memory and attention
• Ingendoh et al. (2023): systematic review of binaural beats and EEG entrainment

This educational analysis is based primarily on the indexed abstract, bibliographic record and publisher preview of the original publication, supplemented by later peer-reviewed reviews. It is not a substitute for the complete paper or professional medical and scientific advice.