MEG and EEG are two powerful, non-invasive techniques used to map the rapid electrical activity within the human brain. Both methods provide a millisecond-by-millisecond view of neural function, making them indispensable for studying cognitive processes and neurological disorders. While they share the goal of understanding brain dynamics, the physical properties they measure and the operational demands they require are vastly different. These distinctions define their respective roles in modern neuroscience and determine why one method might be chosen over the other.
The Physics of Neural Signals
The signals captured by both MEG and EEG originate from the coordinated flow of ions across the membranes of pyramidal neurons in the cortex. This activity, known as postsynaptic potentials, creates tiny electric currents within the brain tissue. When thousands of these neurons fire synchronously, the resulting current is large enough to be detected outside the head.
Electroencephalography measures the electrical potential fields generated by these currents as they spread to the scalp. These currents flow through various tissue layers, including the meninges, the skull, and the scalp. The skull has a high electrical resistance, which significantly blurs the signal, making it challenging to precisely determine the original source location. EEG is sensitive to currents oriented both perpendicular (radial) and parallel (tangential) to the cortical surface, but it favors radial components.
Magnetoencephalography measures the extremely weak magnetic fields that are a byproduct of electric current, as described by the Biot-Savart Law. These magnetic fields are not distorted by the skull and scalp tissues because biological tissues are magnetically transparent. This preservation allows for superior spatial localization compared to EEG. However, MEG is primarily sensitive to tangential currents, which typically arise from neurons located in the sulci, or folds, of the cortex.
Practical Setup and Environment
The operational requirements for MEG and EEG measurements represent a significant practical difference. Electroencephalography is known for its simplicity, portability, and relatively low cost, with systems ranging up to over 256 channels. The measurement involves placing electrodes directly onto the scalp using a conductive gel, and the setup can be easily transported and used in nearly any environment.
Magnetoencephalography is a highly complex and expensive technology requiring specialized, fixed infrastructure. The brain’s magnetic fields are exceptionally weak, typically less than one-billionth the strength of the Earth’s magnetic field. To detect these faint signals, MEG systems historically rely on sensitive Superconducting Quantum Interference Devices (SQUIDs). SQUIDs must be kept immersed in liquid helium at cryogenic temperatures, leading to substantial running costs and the need for recycling systems.
All MEG measurements must be conducted inside a magnetically shielded room (MSR), typically constructed from layers of high-permeability alloys and copper. This MSR is a large, permanent installation necessary to block environmental magnetic noise from sources like traffic and electrical interference. The patient’s head is placed inside a helmet-like array of sensors, although newer systems using optically pumped magnetometers (OPMs) are beginning to mitigate the need for bulky SQUIDs and large MSRs.
Comparing Resolution and Utility
Both MEG and EEG excel in temporal resolution, capturing brain activity on the order of milliseconds. This is necessary to track the rapid dynamics of neural communication and is an advantage over imaging techniques like functional Magnetic Resonance Imaging (fMRI), which measures slower blood flow changes. The methods diverge in their spatial resolution, or the ability to pinpoint the exact location of the activity.
MEG generally offers superior spatial resolution for cortical sources because the magnetic fields are not blurred by the skull. This makes MEG the preferred tool for precise source localization, such as in pre-surgical mapping for epilepsy or tumor removal. Its sensitivity to tangential currents makes it particularly good at mapping activity in the sulci, which constitute a large portion of the cortical surface.
EEG remains the standard for routine clinical applications due to its accessibility and lower cost, despite the blurring effect of the skull. It is widely used for sleep studies, general brain monitoring, and detecting activity from radial sources, which MEG is largely blind to. Combining the two techniques is often considered the most informative approach, as their different sensitivities provide complementary data that improves overall source localization accuracy.