Electrocorticography (ECoG)
Also: ECoG, intracranial EEG (when using surface grids), subdural grid recording
By neuraspeak editorial · Updated 2026-10-07 · 1 min read
Electrocorticography (ECoG) is a method of recording the brain's electrical activity using a thin sheet of electrodes placed directly on the surface of the cortex, under the skull, without penetrating brain tissue.
How does ECoG differ from EEG?
EEG electrodes sit on the scalp, so the signal passes through skull and skin, which blur and weaken it. ECoG electrodes rest on the brain's surface, so each electrode reflects a smaller patch of cortex with much stronger signals. A 2018 review describes intracranial recordings, including ECoG, as resolving neural activity at the millimeter scale in space and the millisecond scale in time (Parvizi and Kastner, Nature Neuroscience). ECoG captures higher-frequency activity, especially the high-gamma band (roughly 70 to 150 Hz), which tracks local neural firing and is hard to measure reliably from the scalp.
The tradeoff is surgery: placing an ECoG array requires opening the skull. Historically, ECoG has been used in hospitals to map seizure activity in people with epilepsy, which is how many early speech-decoding experiments were performed.
How does ECoG differ from intracortical arrays like the Utah array?
ECoG electrodes do not penetrate the brain, so they record the summed activity of many thousands of neurons under each contact rather than the firing of individual neurons. Intracortical arrays insert tiny needles into the cortex and can detect spikes from single neurons or small groups. ECoG can cover a wider area of cortex with one array, while penetrating arrays offer finer detail over a few square millimeters each.
Why it matters for speech BCI
ECoG is one of the two leading recording approaches for speech neuroprostheses. UCSF researchers used high-density ECoG over the speech motor cortex to decode words in a man with anarthria (Moses et al., NEJM 2021), and later used a 253-channel ECoG array to decode text at a median 78 words per minute and drive a synthesized voice and facial avatar (Metzger et al., Nature 2023). Earlier ECoG work in epilepsy patients showed speech could be synthesized from decoded articulatory movements (Anumanchipalli et al., Nature 2019).
Questions
Is ECoG safe?+
ECoG requires neurosurgery and carries surgical risks such as infection and bleeding. It is less invasive than penetrating arrays because it does not enter brain tissue. Long-term safety for chronic BCI use is still being studied in clinical trials.
What is the difference between ECoG and iEEG?+
Intracranial EEG (iEEG) is an umbrella term for recordings from inside the skull. It includes ECoG, which uses surface grids or strips, and stereo-EEG, which uses depth electrodes inserted into the brain.
Related: EEG vs intracortical recordingUtah arraySpeech neuroprosthesisNeural decoderBrain-computer interface (BCI)
Sources
- Parvizi and Kastner, Promises and limitations of human intracranial electroencephalography, Nature Neuroscience 21, 474-483 · 2018
- Moses et al., Neuroprosthesis for Decoding Speech in a Paralyzed Person with Anarthria, New England Journal of Medicine · 2021
- Metzger et al., A high-performance neuroprosthesis for speech decoding and avatar control, Nature 620, 1037-1046 · 2023
- Anumanchipalli, Chartier and Chang, Speech synthesis from neural decoding of spoken sentences, Nature 568, 493-498 · 2019
Facts on this page are drawn from the sources listed and were accurate to the best of our knowledge on the update date shown. Company and product names are trademarks of their owners and are used for identification only. Nothing here is medical, investment or legal advice. Companies and individuals can request a correction via the corrections page.