Inside the Meditating Brain: First Direct Recordings from the Amygdala
For the first time, researchers recorded directly from electrodes deep inside the brain during meditation — revealing gamma power surges in the amygdala and hippocampus.
The Scalp EEG Problem
For decades, meditation neuroscience has relied on scalp EEG and fMRI — technologies that are powerful but fundamentally limited when it comes to deep brain structures. Scalp EEG captures summed electrical activity from the cortical surface, but signals from deep structures like the amygdala and hippocampus are largely attenuated by the skull and intervening tissue. fMRI offers better spatial resolution but measures blood flow changes, not neural activity directly, and operates on a timescale of seconds rather than milliseconds. The regions most relevant to emotional regulation — the amygdala and hippocampus — have remained essentially invisible during meditation in human studies.
An Unprecedented Opportunity
Maher et al. (2025), published in PNAS, found a way around this limitation. Eight patients with drug-resistant epilepsy who had been chronically implanted with NeuroPace Responsive Neurostimulation devices — FDA-approved systems that continuously record intracranial EEG from deep brain structures — participated in a guided loving-kindness meditation session. Six patients had bilateral implantation and two had unilateral (left hemisphere) placement, providing 50 electrode contacts across 14 amygdala channels and 14 hippocampal channels. These patients were all novice meditators with no prior meditation experience. The session consisted of 5 minutes of audio-guided instruction followed by 10 minutes of loving-kindness meditation, guided by materials from the Healthy Minds Program and MBSR resources.
Gamma Surges in the Emotional Core
The recordings revealed a striking pattern. During loving-kindness meditation, gamma power (30-55 Hz) increased significantly in both the amygdala (P < 0.01) and hippocampus (P < 0.001). No other frequency band — delta, theta, alpha, or beta — showed significant power changes. At the individual level, 75% of participants showed increased amygdala gamma and 86% showed increased hippocampal gamma. The study also found that beta oscillatory burst duration decreased in both regions during meditation (P < 0.05), while gamma burst duration increased specifically in the amygdala (P < 0.01) but not the hippocampus. This suggests the amygdala may be a particularly responsive target for meditation-induced neuromodulation.
Why Gamma Matters in the Amygdala
Gamma oscillations in the amygdala are linked to emotional processing and emotional regulation. Prior research has shown that amygdala gamma power correlates with major depressive disorder severity, and that during emotional tasks, gamma oscillations can distinguish between bipolar depression and major depression. The finding that even novice meditators can increase gamma activity in the amygdala through a brief meditation session suggests that meditation may function as a form of targeted, noninvasive neuromodulation — producing the kinds of neural changes that have previously only been achieved through electrical brain stimulation or pharmacology. As senior author Ignacio Saez stated: these brain waves 'are affected in mood disorders like depression and anxiety, so the possibility of being able to willfully control these through meditation is pretty amazing.'
Limitations and What Comes Next
The study has important constraints. Eight participants is a small sample, typical for intracranial research but limiting for generalizability. The participants were epilepsy patients, which may not reflect healthy populations. There was no active control condition beyond the pre-meditation instruction period, and no post-meditation baseline to assess persistence of effects. The 250 Hz sampling rate of the NeuroPace device limited analysis to frequencies below 55 Hz, preventing assessment of broadband high-frequency activity that may also be relevant. Despite these limitations, this study builds directly on the landmark Lutz et al. (2004) work — which found gamma changes in experienced Tibetan monks using scalp EEG — by demonstrating similar effects in novices, using intracranial recordings, in deep limbic structures. Richard J. Davidson, a co-author of both studies, connects the lineage directly.
References
- [1]Maher, C., Tortolero, L., Jun, S., Cummins, D. D., Saad, A., Young, J., Nunez Martinez, L., Schulman, Z., Marcuse, L., Waters, A., Mayberg, H. S., Davidson, R. J., Panov, F., & Saez, I. (2025). Intracranial substrates of meditation-induced neuromodulation in the amygdala and hippocampus. PNAS, 122(6), e2409423122.
- [2]Lutz, A., Greischar, L. L., Rawlings, N. B., Ricard, M., & Davidson, R. J. (2004). Long-term meditators self-induce high-amplitude gamma synchrony during mental practice. PNAS, 101(46), 16369-16373.
- [3]Fell, J., et al. (2010). Medial temporal theta/alpha power enhancement precedes successful memory encoding. Neuroscience, 171(3), 725-730.