Intracranial Validation of Magnetoencephalography Across Oscillatory Frequency and Depth

The human brain is a vast, silent landscape of electrical activity, constantly firing signals that shape our thoughts, memories, and movements. For decades, scientists have relied on tools like functional magnetic resonance imaging, electroencephalography, and magnetoencephalography to peek inside this landscape without opening the skull. Among these, magnetoencephalography stands out because it listens to the faint magnetic whispers produced by electrical currents flowing through brain cells. Unlike other methods that measure blood flow or electrical signals on the scalp, this technique captures the brain's activity at the exact speed it happens, offering a window into the intrinsic rhythm of human cognition. Recently, new sensors have made these recordings possible in more natural settings, allowing researchers to study people while they move or interact with the world. Yet, a fundamental question has lingered: just how much of the brain's true activity can these external sensors actually hear?

To answer this, researchers turned to a rare and powerful opportunity. They worked with a group of patients who were already undergoing treatment for epilepsy, a condition that often requires placing electrodes directly inside the brain to locate the source of seizures. These patients provided a unique chance to record brain activity from two perspectives at once: from the outside using magnetoencephalography and from the inside using the implanted electrodes. By comparing the signals from both sources simultaneously, the team could see exactly how well the external magnetic recordings matched the real neural activity happening deep within the brain. This direct comparison allowed them to test whether the external method was truly capturing the brain's spontaneous rhythms or if it was missing crucial details.

The results revealed a clear picture of what the external sensors can and cannot detect. The study found that when scientists use the external method to map the brain's activity, their estimates of power and sudden bursts of energy accurately reflect what is actually happening inside. This agreement was strongest for slower, lower-frequency rhythms, such as those in the delta, theta, and alpha bands, and for activity occurring near the surface of the brain. As expected, the match was less perfect for the very fast gamma frequencies. However, the most significant discovery was that the external sensors were not limited to the brain's surface. The researchers found that the method was sensitive enough to capture the oscillatory power and burst dynamics of deep structures, specifically the hippocampus, which is vital for memory. This sensitivity to deep brain regions was most robust in the theta frequency range.

These findings confirm that the external magnetic recordings are not just a vague approximation but a precise tool for observing physiologically meaningful activity in both the outer layers and the deeper core of the brain. By establishing that these recordings can reliably track the brain's natural rhythms across different depths and frequencies, the work lays a solid foundation for interpreting future studies. It assures researchers that when they use this technology to explore human cognition in health and disease, they are listening to the brain's true voice, even from the quietest corners of the mind.