The claim that meditation changes the brain has become common enough to feel unremarkable, which is a problem, because the research it rests on is more interesting, more qualified, and less tidy than the headline suggests. There are real neuroimaging findings that have replicated across labs. There are also widely repeated claims that have not held up as well as the popular literature implies. Understanding what the evidence actually shows is more useful than either credulity or dismissal.
The modern research program largely dates to the early 2000s, when functional MRI became widely available and researchers including Richard Davidson at Wisconsin and Sara Lazar at Massachusetts General Hospital began scanning experienced meditators. The findings that have accumulated since then are meaningful, but they need to be read with the same skepticism applied to any young field with a strong publication bias.
What has replicated reasonably well
A few structural and functional findings have appeared across enough independent studies to take seriously.
- Changes in the default-mode network. Experienced meditators show altered activity and connectivity in the network associated with self-referential thought and mind-wandering. This is one of the most consistent findings in the literature.
- Attention network changes. Studies using attention tasks have shown modest improvements in sustained attention and conflict monitoring after meditation training, with corresponding changes in prefrontal and anterior cingulate activity.
- Amygdala reactivity. Several studies have found reduced amygdala response to emotional stimuli after MBSR and related programs, along with structural changes in some samples.
- Insula changes. The insula, involved in interoception, tends to show altered activity and in some studies increased gray matter density in experienced meditators.
These findings are not dramatic in effect size. They are consistent enough to suggest something real is happening.
What has not held up as well
Some of the more widely repeated claims deserve more caution.
- Large gray matter increases from short-term practice. Early studies suggesting substantial structural changes after eight weeks of MBSR have not always replicated at the effect sizes originally reported.
- Telomere and epigenetic effects. These have been reported but rest on small samples and have not yet reached the level of confidence that would justify strong claims.
- Meditation as a general cognitive enhancer. The evidence for broad transfer to unrelated cognitive tasks is weaker than sometimes suggested.
A 2014 meta-analysis in JAMA Internal Medicine, and several subsequent reviews, concluded that the evidence for meditation interventions is strongest for anxiety, depression, and pain, and weaker or absent for many of the other outcomes it has been claimed to improve.
The finding that meditation changes the brain is not particularly informative. Almost everything you do repeatedly changes the brain.
What the mechanism might be
The more interesting question is what specifically is being trained. The current best working model, informed by both cognitive neuroscience and phenomenology, is that meditation practices train a small number of related capacities: sustained attention, meta-awareness (noticing what the mind is doing), and a shift in the relationship to internal experience from identification to observation. The neural correlates of these capacities overlap with the findings above.
This is a more modest claim than meditation rewires your brain, but it is more likely to be true and more useful for deciding whether to practice.
Practical implications
A few things follow from taking this evidence seriously.
- The dose matters. Most of the structural findings come from people with hundreds or thousands of hours of practice. Eight weeks of MBSR is enough to produce measurable clinical effects, but is not enough to remake the brain.
- The type of practice matters. Focused attention, open monitoring, and loving-kindness practices appear to have partially different neural signatures. Choosing a practice and staying with it is more useful than sampling many.
- The effect is cumulative and slow. The changes are more like the changes from consistent exercise than the changes from a medication.
The bottom line
Meditation does produce measurable changes in the brain. The changes are real, modest, cumulative, and specific to what the practice actually trains. They are not magic, and they are not immediate. Understanding the shape of the evidence makes it easier to hold reasonable expectations about what a practice will and will not do.