Abstract
In the first large study of its kind, we quantified changes in electrocorticographic signals associated with motor movement across 22 subjects with subdural electrode arrays placed for identification of seizure foci. Patients underwent a 5-7 d monitoring period with array placement, before seizure focus resection, and during this time they participated in the study. An interval-based motor-repetition task produced consistent and quantifiable spectral shifts that were mapped on a Talairach-standardized template cortex. Maps were created independently for a high-frequency band (HFB) (76-100 Hz) and a low-frequency band (LFB) (8-32 Hz) for several different movement modalities in each subject. The power in relevant electrodes consistently decreased in the LFB with movement, whereas the power in the HFB consistently increased. In addition, the HFB changes were more focal than the LFB changes. Sites of power changes corresponded to stereotactic locations in sensorimotor cortex and to the results of individual clinical electrical cortical mapping. Sensorimotor representation was found to be somatotopic, localized in stereotactic space to rolandic cortex, and typically followed the classic homunculus with limited extrarolandic representation.
Introduction
Functional brain mapping—identifying which parts of the brain control specific body parts—is critical for both neurosurgery and developing brain-computer interfaces (BCIs). Traditionally, researchers used EEG (which is non-invasive but "blurry" due to the skull) or fMRI (which is precise but slow). While Electrocorticography (ECoG)—recording directly from the brain's surface—offers high spatial and temporal resolution, early studies were often limited to small case reports. This 2007 study by Miller et al. represents a landmark effort in human neuroscience, providing one of the first large-scale, standardized looks at how our brains' electrical "rhythms" change when we move.
Methods
The researchers studied 22 epilepsy patients who had electrode arrays temporarily implanted to locate seizure foci. This allowed for direct recording from the cortical surface with high fidelity.
Patients performed simple, repetitive movements (like clenching a fist or protruding the tongue) in timed intervals. Signal analysis focused on two specific spectral bands: a Low-Frequency Band (LFB) of 8–32 Hz and a High-Frequency Band (HFB) of 76–100 Hz. To compare data across 22 different brains, electrode locations were mapped onto a standardized "Talairach" template cortex, allowing for group-level averaging.
Key Insights: The Spectral Tilt and Somatotopy
Movement causes a consistent "spectral tilt" in the ECoG signal. Power in the low frequencies (LFB) decreases (event-related desynchronization), while power in the high frequencies (HFB) increases.
A critical finding was the difference in spatial distribution between these bands. The HFB increase is highly focal, occurring in specific, small patches of cortex. In contrast, the LFB decrease is spatially broad. This suggests that while LFB might represent a general "ready" state of a brain circuit, HFB represents the actual computation of the movement. Furthermore, the HFB maps followed the classic homunculus with remarkable accuracy, localized primarily to the rolandic cortex (Brodmann area 4).
Results and Validation
The study found high consistency across subjects: LFB power decreased in 17 of 18 subjects during hand tasks, while HFB power increased in 15 of 18. The average activation locations for hand and tongue movements fell exactly within the expected stereotactic coordinates for the motor cortex. In seven patients, these passive recordings were compared to active Cortical Stimulation Mapping; in 16 out of 20 electrode pairs tested, the spectral changes correctly predicted the results of clinical electrical stimulation. Statistical analysis confirmed that HFB changes were significantly more spatially constrained than LFB changes ($p < 0.01$).
Limitations and Future Directions
Electrode placement was determined by clinical necessity for epilepsy treatment, which limited coverage in some subjects. Additionally, the mapping technique used Gaussian kernels to smooth data, which might merge distinct adjacent active sites. However, because HFB changes are so robust and focal, they are ideal candidates for controlling brain-computer interfaces, potentially allowing for real-time control of prosthetic limbs using high-frequency cortical signals.
Conclusion
By demonstrating that high-frequency brain activity provides a focal, somatotopic map of movement, this work validates the use of HFB spectral shifts as a robust tool for functional brain mapping and neural prosthetics. The discovery of the focal nature of HFB versus the broad nature of LFB provides a new framework for understanding cortical dynamics during human motor behavior.
Summary updated