Episode Summary
I sat down with the team at "Groundbreaking Experiment" to run an unusual test. They brought a practicing medium, Susan, to Peak Brain Institute in Los Angeles and asked a simple question: what does the brain actually do during a mediumship reading, and can an EEG capture it? You can watch the original conversation. Here is the neuroscience I walked them through, and what I would tell you if you brought me the same question.
What does a QEEG brain map actually measure?
An EEG records the summed electrical activity of large populations of cortical neurons through electrodes on the scalp. A QEEG, or quantitative EEG, takes that raw signal and compares it against a normative database so you can see where a given brain sits relative to thousands of others at each location and each frequency band. I have collected and read more than 25,000 of these maps. If you want the full walkthrough of the procedure, I lay it out in the QEEG brain mapping guide.
A QEEG measures rhythmic electrical activity at the scalp. It tells you about the relative balance of delta, theta, alpha, beta, and gamma at each region, and it tells you about timing and connectivity between regions. It does not read the content of a thought. It cannot tell you what someone is thinking, whether a belief is true, or where information is coming from. When I map a brain, I am looking at the operating characteristics of the cortex, not at the meaning a person assigns to their experience.
That distinction matters for an experiment like this one. The EEG can show us the state Susan goes into. It cannot adjudicate the source of what she reports.
What happens in the brain during a focused, internally-directed state?
A mediumship reading is, at the level of brain activity, an internally-generated attentional state. The person closes off external input and attends to imagery, felt sense, and spontaneous verbal material arising from inside. We have decent markers for that kind of state.
When attention turns inward and external sensory processing quiets, alpha power tends to rise, especially over posterior and occipital regions. Alpha, the 8 to 12 Hz rhythm, functions as a cortical idle and a gate. It rises when a region is taken offline from active processing. I walk through this in detail in Decoding Alpha Waves. High posterior alpha with eyes open is one signature of someone who has pulled attention away from the room and turned it inward.
You also tend to see shifts in frontal midline theta during sustained internally-focused attention. Frontal midline theta, around 4 to 7 Hz over the anterior cingulate region, ramps up during concentrated effort and during certain meditative and absorbed states. It reflects the prefrontal cortex and anterior cingulate coordinating sustained attention.
States of deep absorption, the kind reported during trance, flow, and some contemplative practice, share a family of EEG features: elevated frontal theta, shifts in alpha distribution, and changes in connectivity between frontal and posterior regions. I cover the absorbed end of this in Biohacking Flow State and the contemplative end in Biohacking Meditation. The same machinery that lets a meditator drop into a deep practice state is the machinery a medium recruits during a reading. The brain recruits attention circuits, default-mode circuits, and sensory-gating circuits, and reconfigures how they talk to each other. There is no separate "psychic" circuit in the map.
Can an EEG prove or disprove mediumship?
An EEG can confirm that a person entered an unusual, internally-directed state. It can show that the state is reliable and reproducible across sessions. It can distinguish that state from ordinary alert wakefulness or from someone faking concentration. Identifying the source of the information a medium reports sits outside what the instrument can do.
This is a measurement-scope problem. The signal I record is electrical activity from the person's own cortex. If Susan reports a name or an image, the EEG tells me her brain generated that report and what state it generated it in. The instrument is silent on whether the content came from memory, from inference, from chance, or from somewhere it was never built to detect. Conflating "the brain did something measurable" with "the claimed source is real" is the error I work hardest to avoid. The first is well-established. The second sits entirely outside what scalp EEG can address.
I apply the same discipline to clinical work. When I read a map for ADHD or anxiety, I report what the rhythms show and what training can change. Overreaching into claims the data cannot support is where EEG interpretation goes wrong, regardless of the application.
Why is everyone's "absorbed" brain a little different?
People assume a given mental state should look identical across brains. Repeated measurement across different people in the same state shows something else: recognizable family features alongside unique individual signatures. Two people doing focused internal attention produce different alpha distributions, different theta dynamics, different connectivity patterns. The state is identifiable as a category; the individual expression of it varies. This is the core of what I call EEG phenotyping, and I unpack it in Biohacking with EEG Phenotypes.
For Susan, repeated measurement across multiple sessions is the scientifically useful move. A consistent, identifiable EEG signature across sessions is itself a finding. It tells you the state is a real, reproducible brain configuration rather than a one-off. Reproducibility is the bar I hold any pattern to before I treat it as meaningful.
Could a medium train this state on purpose?
The state markers I described respond to operant conditioning. Neurofeedback works by feeding back a real-time signal from your own EEG and rewarding shifts toward a target pattern. Over repeated sessions the brain learns to produce the target state more readily and to hold it longer. If you want the mechanism and the evidence base, I cover it in Is Neurofeedback Legitimate? and across the neurofeedback topic page.
A practitioner who relies on entering a specific absorbed state could, in principle, train that state with feedback the same way an athlete trains a focus state or a person with anxiety trains down an over-active threat response. Train alpha for the disengagement from external input, train frontal theta for the sustained internal attention, and entry into the state becomes faster and more reliable. I have not run that protocol with a medium, so this is extrapolation from how the relevant circuits respond in other contexts, not a published finding. The mechanism is sound; the specific application is untested.
What I took away from the experiment
Bringing instruments to a question most people treat as off-limits is good practice, and that is what the "Groundbreaking Experiment" team set out to do. The EEG gave us something concrete: a measurable, internally-directed brain state during the reading, distinct from ordinary waking attention. That is a real result, and it is the result the instrument is qualified to give.
The map shows the state and is silent on the source. If you want to study a practice like this seriously, run repeated sessions, characterize the EEG signature across time, and keep the claim attached to what the data can carry. The open question about source stays open until someone designs an instrument actually capable of addressing it.