What does it feel like to act through a brain-computer interface?
People with paralysis who used brain-computer interfaces felt like the authors of the device's actions and valued the social role it gave them, but were unsure who was to blame when it failed.
Source
What is it like to use a BCI? - insights from an interview study with brain-computer interface users
Study at a glance
- Design
- Qualitative / archival — Semi-structured interviews with BCI users, analysed with Grounded Theory open coding into categories
- N
- N=9 · 9 BCI users with physical impairments who had taken part in experimental BCI studies; one failed the test session
- Population
- Adults aged 24 to 77 with muscular conditions or paralysis from Germany, the US and France who had used active or reactive BCIs
- Outcome
- How users experience agency, social participation and self-definition when acting through a BCI
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
The researchers interviewed nine people with physical impairments who had used brain-computer interfaces (BCIs) in research, some with electrodes implanted in the motor cortex and others using EEG caps. Interviews covered daily life, first BCI experiences, what acting via a BCI felt like, and hopes and risks. Transcripts were coded with a Grounded Theory approach into overarching categories.
What they found
Three categories emerged: being an agent, participation and self-definition. Users who operated a BCI successfully felt they were the agent when the output matched their intention, and felt responsible for successes, but shifted blame between themselves and the technology after errors. Many valued being part of a research team, public recognition and meaningful occupation more than the BCI tasks themselves. Self-definition varied widely, from treating the BCI as a mere tool to incorporating a robotic arm as 'my arm' or centering one's identity on the brain; users also reported having to suppress emotions because feelings disrupted control.
The limits
What it doesn't show
The sample is very small and heterogeneous in diagnosis, BCI type and experience, and recruitment through gatekeepers and existing participants may favor enthusiastic users. Only research participants were interviewed, so the social benefits may reflect the research setting rather than BCIs as everyday devices. Other stakeholders such as caregivers, developers and clinicians were not included. The findings describe how users feel; they do not settle who is actually morally responsible for BCI-mediated actions.
Key terms
- Brain-computer interface (BCI)
- A device that records brain activity and converts it into commands for computers, prostheses or other outputs.
- Active vs reactive BCI
- Active BCIs use a mental task like imagining movement; reactive BCIs use the brain's response to attending to an external stimulus.
- Sense of agency
- The feeling that I am the one causing or generating an action.
- Illusion of control
- The tendency to overestimate one's control, for example claiming successes while blaming failures on outside causes.
- Distributed agency
- The idea that in advanced technologies human and machine contributions are so intertwined that no single agent produces the outcome.
- Brainification
- The authors' term for placing the brain at the center of one's self-image, echoing a neuroscience worldview.
Flashcards
0 of 11 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
Which analysis method did the study use?
Common questions
Is this philosophy or psychology?
It is empirical neuroethics: the interviews are used to inform philosophical questions about agency, responsibility and personal identity with neurotechnology.
Did users feel less human or like cyborgs?
No. None described themselves as hybrids; most saw technology as a way to regain independence, and one said she remained the same person.
Why does the paper matter for responsibility debates?
Users claimed successes but hesitated over failures, suggesting single-agent models of responsibility may fit BCI actions poorly and that distributed agency may be needed.
More on Neuroethics