Brain-computer interface (BCI): definition, types and how it differs from organoid intelligence

A brain-computer interface (BCI) is a system that records a person’s brain activity and translates it into commands for a computer or device, bypassing the usual pathways of nerves and muscles. BCIs range from non-invasive EEG caps to implanted electrode arrays, and mostly aim to restore communication and movement. Unlike organoid intelligence, a BCI connects to a living person’s brain.

Key takeaways

• Wolpaw et al. (2002) defined a BCI as a communication and control system that does not rely on the brain’s normal output through peripheral nerves and muscles.
• BCIs are grouped by invasiveness: scalp EEG, surface ECoG and intracortical microelectrode arrays.
• Implanted BCIs have let people with paralysis control cursors and, in Willett et al. (2021), write text at about 90 characters per minute.
• BCIs interface with a person’s brain; organoid intelligence uses cultured neural tissue grown in a lab as a computing substrate.
• The fields share hardware, decoding methods and closed-loop software, but differ sharply in goals, risks and ethics.

What is a brain-computer interface? #

Normally, when you decide to move your hand, signals travel from the brain through the spinal cord and nerves to the muscles. Injury or disease can break that chain while leaving the brain’s intentions intact. A BCI creates a new output path: it measures brain activity directly, works out what the person is trying to do, and sends that to a device.

In their widely cited review in Clinical Neurophysiology (2002), Wolpaw, Birbaumer, McFarland, Pfurtscheller and Vaughan defined a BCI as a communication system in which messages or commands do not pass through the brain’s normal output pathways of peripheral nerves and muscles. They emphasized that a BCI is an interaction between two adaptive systems: the user’s brain, which learns to produce useful signals, and the software, which learns to interpret them.

How does a brain-computer interface work? #

Most BCIs follow the same basic loop:

1. Signal acquisition. Sensors record brain activity, for example EEG from the scalp or spikes and field potentials from implanted electrodes.
2. Preprocessing. Filtering removes noise and separates useful frequency bands or spike events.
3. Feature extraction. Software computes features such as firing rates of recorded neurons or power in particular rhythms.
4. Decoding. A model, often trained on calibration data, maps features to an intended output, such as cursor velocity or a selected letter.
5. Device output. The decoded command drives a cursor, speller, robotic arm or other device.
6. Feedback. The user sees the result and adjusts, closing the loop.

Steps 3 and 4 are the domain of neural encoding and decoding. Some BCIs also write information back into the nervous system through stimulation, making them bidirectional, which connects them to closed-loop stimulation research.

What are the main types of brain-computer interfaces? #

The physics of these signals is the same electrophysiology used in the lab. Intracortical BCIs use implanted arrays that are close relatives of the in vitro microelectrode array used with cultured neurons, though they differ in shape, materials and safety requirements.

What have brain-computer interfaces achieved? #

BCI research has moved from laboratory demonstrations to studies with human participants. Two milestones illustrate the progress:

• Cursor and device control. Hochberg et al. reported in Nature (2006) that a person with tetraplegia used an implanted microelectrode array in motor cortex to control a computer cursor and other devices with neural activity.
• Brain-to-text communication. Willett, Avansino, Hochberg, Henderson and Shenoy reported in Nature (2021) that a participant with paralysis, imagining handwriting, produced text at about 90 characters per minute, with a decoder translating neural activity from implanted arrays into letters.

These are research systems used under clinical study conditions, with close support from teams of scientists and clinicians. Turning them into reliable, long-lasting, everyday devices remains an active challenge.

How does a brain-computer interface differ from organoid intelligence? #

People sometimes group BCIs and organoid intelligence together because both involve neurons, electrodes and computers. They are different fields with different aims.

The simplest way to remember it: a BCI helps a brain use a computer; organoid intelligence tries to use neurons as part of the computer. Smirnova et al. (Frontiers in Science, 2023) set out organoid intelligence as a form of biocomputing built on brain organoids and cultured neural tissue. Experiments such as DishBrain, in which Kagan et al. (Neuron, 2022) embedded cultured neurons in a simulated game, involve no person’s brain at all.

Where the two fields overlap #

Despite different goals, the fields share technical problems:

• Recording stability. Electrodes must work reliably for long periods, whether implanted or under a culture.
• Real-time decoding. Both turn noisy neural activity into useful outputs within milliseconds.
• Stimulation design. Both need safe, informative ways to send signals into neural tissue.
• Adaptation. In both, the neurons and the software adapt to each other, so decoders may need updating over time.
• Data standards. Both generate large recordings that benefit from shared formats such as Neurodata Without Borders.

Our article on bio-silicon hybrid computing explores where living neurons and chips meet across these fields.

What are the ethical issues with brain-computer interfaces? #

BCIs raise questions specific to connecting technology to a person’s brain:

• Safety. Implanted devices involve surgical risk and long-term questions about tissue response.
• Consent and autonomy. Participants, some of whom have severe paralysis, must be able to give and withdraw informed consent.
• Neural data privacy. Recorded brain activity is sensitive personal data, and its storage and use need protection.
• Agency. When a decoder misreads intent, it can be unclear who is responsible for the resulting action.
• Access and support. Users may come to depend on devices that need ongoing technical and clinical support.

Organoid intelligence has a different ethical profile, centered on cell donor consent, provenance and the status of cultured tissue. Our article on the ethics of organoid intelligence covers those questions.

Why does the distinction matter? #

Clear language helps the public, investors and regulators understand what each field does. A BCI is a medical and assistive technology for people. Organoid intelligence is laboratory research into computing with living neural tissue. At Organoid Programming, we work on the second: software for designing and reproducing experiments on brain organoids, with consent and provenance attached. For a full introduction, see what is organoid intelligence. Understanding the brain-computer interface, and how it differs, is a useful starting point for anyone new to neurotechnology.

Frequently asked questions

What is a brain-computer interface in simple terms?

A brain-computer interface is a system that lets a person control a computer or device using brain activity alone. Sensors record signals from the brain, software decodes what the person intends, such as moving a cursor or writing a letter, and the device carries it out. Wolpaw et al. (2002) defined BCIs as systems that do not depend on the brain’s normal output through nerves and muscles.

What are the main types of brain-computer interfaces?

BCIs are usually grouped by how close the sensors are to the brain. Non-invasive BCIs, most commonly EEG, record from the scalp. Partially invasive approaches such as ECoG place electrodes on the brain’s surface. Intracortical BCIs implant microelectrode arrays into brain tissue to record individual neurons. Closer sensors give richer signals but require surgery and carry more risk.

How is a brain-computer interface different from organoid intelligence?

A BCI connects a computer to a living person’s brain, usually to restore communication or movement, and the person stays in control of what they intend. Organoid intelligence uses brain organoids, small 3D cultures of neurons grown from stem cells in a lab, as a computing substrate. No person’s brain is connected. The two fields share tools such as electrodes and decoding algorithms but have different goals and ethics.

What can brain-computer interfaces do today?

Research BCIs have allowed people with paralysis to control computer cursors and robotic devices and to communicate. Hochberg et al. reported in Nature (2006) that a person with tetraplegia used an implanted array to control a cursor. Willett et al. reported in Nature (2021) that a participant with paralysis produced text at about 90 characters per minute by attempting to handwrite letters.

Do brain-computer interfaces read thoughts?

Not in the everyday sense. Current BCIs decode specific, trained signals, such as attempted movements or particular brain responses, from the regions where sensors are placed. They require calibration and typically work only for the task they were built for. Still, neural data can be sensitive, so privacy, consent and data protection are important ethical concerns in BCI research.

Could organoid research help brain-computer interfaces?

Possibly, indirectly. Both fields face similar engineering problems: recording reliably from neurons, decoding activity in real time, stimulating safely, and keeping electrodes working for long periods. Advances in microelectrode arrays, closed-loop software and decoding methods developed for one field may inform the other. Organoids are also studied as models of human neural tissue, though they are very different from an adult brain.

Sources

1. Wolpaw, Birbaumer, McFarland, Pfurtscheller & Vaughan (2002). Brain-computer interfaces for communication and control. Clinical Neurophysiology.
2. Hochberg et al. (2006). Neuronal ensemble control of prosthetic devices by a human with tetraplegia. Nature.
3. Willett, Avansino, Hochberg, Henderson & Shenoy (2021). High-performance brain-to-text communication via handwriting. Nature.
4. Smirnova et al. (2023). Organoid intelligence (OI): the new frontier in biocomputing and intelligence-in-a-dish. Frontiers in Science.
5. Kagan et al. (2022). In vitro neurons learn and exhibit sentience when embodied in a simulated game-world. Neuron.