Meditation is one of the oldest self-improvement practices human beings do. It is also one of the most-studied. If you look at the meditation literature honestly, four techniques stand out for having accumulated meaningful randomized trial evidence: mindfulness (particularly MBSR), Transcendental Meditation, body scan practice, and loving-kindness (metta) meditation. Everything else is either a variant of one of the four or has thin evidence outside subjective testimony.
Real evidence in this context does not mean guaranteed benefit. It means enough well-designed trials to draw modest conclusions. The most-cited synthesis — Goyal and colleagues (2014, JAMA Internal Medicine) — makes those conclusions precise, and they are more restrained than the wellness industry usually presents. This article walks through what each of the four techniques is, what the Goyal meta-analysis actually concluded, why 10 minutes a day is the right dose to run a personal test at, and how a biofeedback device can help you build the habit if you want the objective input.
The four techniques with real evidence
Mindfulness (MBSR and MBCT)
Mindfulness-based stress reduction (MBSR) was developed by Jon Kabat-Zinn at the University of Massachusetts Medical Center starting in 1979. The formal protocol is an 8-week program of weekly 2.5-hour classes plus daily home practice, incorporating body scan, sitting meditation with breath focus, and gentle yoga. Mindfulness-based cognitive therapy (MBCT) is a related 8-week program adapted specifically for recurrent depression.
The core skill is non-judgmental attention to present-moment experience. In practice, that usually means sitting with the breath as an anchor and returning attention to it when the mind wanders — noticing the wandering itself as part of the practice rather than as failure.
MBSR is the most-researched meditation protocol in the biomedical literature. It has the largest number of trials, the widest range of conditions studied, and the most systematic reviews.
Transcendental Meditation (TM)
Transcendental Meditation was popularized by Maharishi Mahesh Yogi in the 1960s. The formal practice involves silently repeating a personal mantra — assigned by a certified TM teacher — for 20 minutes twice a day, in a specific effortless way that is not concentration in the mindfulness sense. TM is proprietary and taught commercially; instruction costs several hundred to over a thousand dollars in the U.S.
TM has accumulated meaningful cardiovascular evidence. Schneider and colleagues (2012, Circulation: Cardiovascular Quality and Outcomes) published a randomized trial in Black adults with coronary heart disease showing that TM practice over an average of 5.4 years was associated with a 48 percent reduction in the composite endpoint of death, myocardial infarction, and stroke compared to a health education control. That is a large effect from a real trial. Blood pressure reduction and stress-hormone changes have been reported across multiple TM trials.
The TM evidence base is real but comes with a caveat: much of it is produced by researchers affiliated with the TM organization, which raises legitimate questions about publication bias and independent replication. The cardiovascular findings have been replicated across multiple research groups, which strengthens the case, but the field is not fully independent.
Body scan
Body scan meditation involves systematically directing attention to different regions of the body, noticing sensations without trying to change them. It is a core component of MBSR and traces its structure to Vipassana practice, particularly the Goenka lineage.
As a standalone technique, body scan has been studied less than mindfulness as a whole, but it appears in many MBSR trials as one of the practices participants report finding most useful. The mechanism appears to combine attention training with a form of exposure to internal sensations, which may explain its role in trials of chronic pain and interoception-related conditions.
Loving-kindness (metta)
Loving-kindness meditation, or metta, is a practice from the Buddhist tradition in which the practitioner silently repeats phrases of well-wishing directed at themselves, then loved ones, then neutral people, then difficult people, and finally all beings. The classical phrases translate roughly to "may they be happy, may they be healthy, may they be safe, may they live with ease."
Fredrickson and colleagues (2008, Journal of Personality and Social Psychology) published a well-cited randomized trial of loving-kindness meditation showing increases in daily positive emotions and, over time, increases in a range of personal resources (mindfulness, sense of purpose, social support). The effect sizes were modest but consistent, and the mechanism — training positive-affect circuits through directed practice — has a defensible neurobiological rationale.
Metta has emerged as a distinct evidence-supported technique separate from breath-focused mindfulness. Trials have shown effects on empathy, self-compassion, and reduced implicit bias, though effect sizes are typically small.
What Goyal 2014 actually said
Goyal and colleagues (2014, JAMA Internal Medicine) performed a systematic review and meta-analysis of meditation programs for psychological stress and well-being. They included 47 randomized trials with 3,515 participants that used an active control (not just a waitlist). The active-control criterion is important — many meditation trials compare a full 8-week program with home practice to nothing at all, which stacks the deck for the meditation arm.
The findings, honestly stated:
- Moderate evidence that mindfulness meditation programs improve anxiety, depression, and pain.
- Low evidence of improvement in stress and mental health-related quality of life.
- Low or insufficient evidence of any effect on positive mood, attention, substance use, eating habits, sleep, or weight.
- No evidence that meditation was better than active treatments like drugs or exercise for the outcomes it did improve.
This is a real endorsement of meditation for anxiety, depression, and pain — three of the most common reasons people take up a practice. It is also a much more measured endorsement than the wellness industry provides. Meditation is not a general-purpose optimization tool with proven benefits across the board. It is a specific-purpose tool with moderate evidence for a small number of outcomes, and it is not superior to established treatments like antidepressant medication or aerobic exercise for those same outcomes.
Since 2014, additional evidence has accumulated. Goldberg and colleagues (2018) published a larger meta-analysis with generally consistent conclusions and modest effect sizes across a broader range of outcomes. The general shape of the evidence base has not changed: modest, real, worth pursuing, not a miracle.
Why 10 minutes a day is the right dose to test with
Formal MBSR requires 45 minutes of home practice daily for 8 weeks. That is a real commitment and one many beginners will not sustain. The good news is that the practical minimum effective dose appears to be much lower.
Multiple studies — including Balban 2023 (which used 5-minute daily breathwork or meditation practices over 30 days) — have shown measurable effects on mood and anxiety at short daily doses. Basso and colleagues (2019, Behavioural Brain Research) reported meaningful attention and mood improvements with 13 minutes of daily meditation over 8 weeks in previously non-meditating adults. The dose-response for beginners is probably front-loaded: the difference between 0 and 10 minutes a day is much larger than the difference between 10 and 30 minutes a day.
Ten minutes a day is the right dose for a beginner running a personal test for four reasons:
- Sustainable. A 10-minute daily practice is achievable for most people without restructuring the day.
- Enough for signal. The trial evidence suggests 10 to 20 minutes daily is well within the dose that has produced measurable effects.
- Long enough to test. Eight weeks at 10 minutes a day is 4.7 hours of practice. That is enough to detect a meaningful subjective change if there is one to detect.
- Below the sunk-cost trap. A 10-minute daily test is easy to abandon if it is not working. A 45-minute daily commitment attracts sunk-cost reasoning and stubbornness.
How to run the test
- Week 1 to 2 — pick one technique. For most people starting out, mindfulness of breath is the default. Sit comfortably, follow the sensation of the breath, gently return attention when it wanders. Ten minutes, once a day. If you have a specific interest in self-compassion, empathy, or difficult interpersonal work, try loving-kindness. If you have chronic pain or difficulty tolerating internal sensations, try body scan. If cardiovascular outcomes are the goal and you can afford the instruction fee, TM is a defensible alternative.
- Week 3 to 4 — track subjective state. Rate mood, sleep quality, and self-reported stress on a 1-to-10 scale at the same time each day. Do this before you sit and again 30 minutes after. You are looking for both immediate acute effects and multi-week trend changes.
- Week 5 to 6 — hold the practice, evaluate. By this point, the acute effect should be visible if it is going to be. If sitting is producing a measurable shift and the daily trend on mood or anxiety has improved, the practice is earning its place.
- Week 7 to 8 — decide. Compare your week 7 to 8 mood and stress ratings to your week 1 to 2 baseline. If there is a real difference, you have found a tool worth keeping. If there is not, the honest read is that meditation may not be your intervention, or the specific technique may not be the right fit — try a different one for another eight weeks or accept that this is not your tool.
Biofeedback as an accountability tool
You do not need a device to meditate. Human beings have been doing it for millennia without one. But for some people — particularly those who value objective input, who struggle with consistency, or who want a signal that the practice is producing measurable state changes — an EEG-based biofeedback device can help make the practice feel real and support daily consistency.
The honest limit of biofeedback in meditation: it is a scaffolding tool, not the practice. The evidence for the underlying meditation techniques does not require or depend on a device. Muse 2 is useful primarily as an accountability tool and as a way to make abstract mental states legible in the early weeks when it is easy to wonder whether anything is happening. Once a practice is established, most long-term meditators drop the device.
Honest limits and cautions
Meditation is generally safe, but it is not universally so. A few caveats worth naming:
- Meditation is not a substitute for treatment of clinical depression, anxiety, PTSD, or psychosis. It can be a useful adjunct. It is not a primary intervention, and the Goyal 2014 review explicitly notes it is not superior to established treatments.
- A small minority of practitioners experience adverse effects. Britton and colleagues have documented cases of persistent distress, dissociation, or disruption after intensive meditation retreats. Beginners doing 10 to 20 minutes a day are at very low risk; people considering multi-day silent retreats should be aware that the intensity produces occasional negative effects.
- Do not expect meditation to fix a life producing chronic stress. Meditation helps a nervous system meet stress with more equanimity. It does not fix a work situation that is unsustainable, a relationship that is corrosive, or a sleep debt that is chronic. Those problems have their own solutions.
The honest framing: four meditation techniques — mindfulness, TM, body scan, and loving-kindness — have accumulated real evidence for specific outcomes. The evidence is modest, moderate for anxiety and depression and pain, weaker for the more general well-being claims, and consistently unable to beat established treatments head-to-head. Ten minutes a day is the right dose for a beginner to test whether the practice is their tool. Biofeedback can help with consistency but is not required. An 8-week test at 10 minutes a day is enough to make an honest personal decision. That test is worth running.
