Section 391 of 440
Complete canonical tutorial. This reader section contains the same teaching body as PWR-209 · Brain–computer communication. Open the Power dossier.
PWR-209 · SUPERVISED full tutorial
Use a provider-configured brain–computer communication channel while preserving confirmed authorship and a conventional fallback
Alex uses a provider-configured BCI to select YES, NO, PAUSE or HELP, but no decoder output becomes a message until Alex confirms it through the established eye code. The team rejects artefact, repairs a false HELP and keeps the alphabet board ready. These results belong to the complete person–device–decoder–support arrangement; they are not thought reading.
1 · Permission and limits
Know exactly what you may do
2 · Get ready
Gather what you need and check the starting conditions
What you need
- Declared Brain–computer communication fixture: In a supervised session, Alex uses the approved BCI to choose YES, NO, PAUSE or HELP. Alex’s established eye-coded yes/no and alphabet board remain available.
- Setup aid for Check the complete system: Verify approved equipment, decoder/version, trained team, privacy controls and immediate fallback.
- Brain–computer communication log: Person-confirmed message accuracy, rate, correction, calibration/support time, home uptime, privacy and safe fallback versus best alternative access; retain Brain–computer communication errors, assistance, stop and fallback.
- BCI communication provider table: from “A high-performance speech neuroprosthesis”, extract participant, implant, decoder, output, confirmation and measured communication rate; from the long-term independent-use study, extract uptime, support, failure and alternative-access context. Use the NIST Privacy Framework for signal, message and account handling. Neither paper permits unconfirmed decoder output to stand as Alex’s speech.
Before you start
- Obtain protocol or clinical clearance, informed consent and confirmation through an established communication method.
- Verify approved equipment, decoder/version, trained team, privacy controls and immediate fallback.
- Start check for Brain–computer communication: Goal and consent are confirmed independently of the BCI.
- Top-of-sheet stop for Brain–computer communication: Stop for pain, headache, skin change, seizure-like activity, distress, marked fatigue, new neurological symptom or participant withdrawal.
3 · The method
Follow these steps in order
- Set a person-chosen communication goal
Alex confirms the four-message task, acceptable effort, privacy boundary and preferred fallback through the established channel.
Why: Goal and consent are confirmed independently of the BCI.
Check: Goal and consent are confirmed independently of the BCI.
- Check the complete system
The team verifies approved hardware, sensors or implant interface, decoder version, account, display, positioning, skin or clinical status and emergency procedure.
Why: Configuration and responsible staff are logged before acquisition.
Check: Configuration and responsible staff are logged before acquisition.
- Record a no-command state
The provider captures rest or no-selection data and checks noise, artefact and unintended output.
Why: A no-command baseline reveals unintended selections before any message is treated as the person’s communication.
Check: During declared no-command trials, unintended output remains within the provider’s predeclared limit.
- Calibrate the small choice set
The team presents labelled YES, NO, PAUSE and HELP targets with rest breaks and fits only the approved decoder.
Why: Each class has enough valid provider-accepted trials and rejected artefacts are counted.
Check: Each class has enough valid provider-accepted trials and rejected artefacts are counted.
- Send and confirm messages
Alex selects a target; the partner reads it back through the established channel and Alex confirms or rejects before anyone acts.
Why: Message meaning is person-confirmed, not decoder-assumed.
Check: Message meaning is person-confirmed, not decoder-assumed.
- Repair false selection
On an incorrect HELP output, Alex uses PAUSE or the conventional channel, the partner cancels the message and the error enters the log.
Why: Wrong output produces no external action and authorship is restored.
Check: Wrong output produces no external action and authorship is restored.
- Compare with best alternative
Measure confirmed accuracy, rate, correction time, fatigue, setup burden and privacy beside the eye-coded or board method.
Why: The BCI is not judged by raw speed alone.
Check: The BCI is not judged by raw speed alone.
- Finish with fallback and review
The provider stages decoder unavailability; Alex communicates one message through fallback, then the team checks symptoms and makes the next decision.
Why: Fallback works and progression remains team- and participant-controlled.
Check: Fallback works and progression remains team- and participant-controlled.
4 · Worked example
See the whole method used once
Scenario
In a supervised session, Alex uses the approved BCI to choose YES, NO, PAUSE or HELP. Alex’s established eye-coded yes/no and alphabet board remain available.
Walkthrough
- Alex confirms the goal and privacy boundary using the eye-coded channel before BCI setup.
- The team logs the approved interface and decoder version and records stable no-command trials.
- Alex calibrates YES, NO, PAUSE and HELP with scheduled rests; two artefact trials are rejected rather than labelled.
- In ten prompted messages, eight are correct; one false HELP is cancelled after Alex selects PAUSE and rejects it by eye code, while one no-selection trial is completed through the alphabet-board fallback.
- The team records accuracy, correction time, fatigue 2/4 and 25 minutes of setup, then compares with the alphabet board.
- When the decoder is disabled, Alex sends NO through eye code; the team ends the session without adding classes.
Result
Alex confirms eight messages, repairs a false output and uses fallback. The result belongs to this person–device–decoder–support configuration and is not thought reading.
5 · Right and wrong
Compare correct or safer execution with the common wrong version
| Moment | Right / safer | Wrong / riskier | Why it matters |
|---|---|---|---|
| Authorship in four-message BCI session | Confirm every consequential message with the person’s reliable channel. | Treat decoder output as the person’s intention automatically. | Classification errors can misattribute meaning; this distorts the four-message BCI session. |
| Choice set | Begin with a small labelled vocabulary. | Add many commands after one accurate block. | More classes can raise calibration burden and error. |
| Fatigue in four-message BCI session | Schedule rest and record setup/support time. | Count only messages per minute during the four-message BCI session. | High rate can conceal fatigue, artefact or care-partner burden. |
| Fallback in four-message BCI session | Keep the established communication system ready. | Remove it to prove BCI independence. | Device, decoder or connection loss can remove communication. |
6 · Common mistakes
Spot the error and apply the correction
| Mistake | Fix |
|---|---|
| Treat decoder output as the person’s intention automatically. | Use read-back plus accept/reject before action. |
| Add many commands after one accurate block. | Progress only on the provider’s criterion and Alex’s choice. |
| Count only messages per minute. | Score accuracy, correction, fatigue and support together. |
| Remove it to prove BCI independence. | Practise immediate switch to the best alternative. |
7 · Practice
Turn the steps into a usable skill
First session
- Four-message BCI visit: In a supervised session, Alex uses the approved BCI to choose YES, NO, PAUSE or HELP. Alex’s established eye-coded yes/no and alphabet board remain available.
- Interface and decoder record: Check the complete system: The team verifies approved hardware, sensors or implant interface, decoder version, account, display, positioning, skin or clinical status and emergency procedure.
- Provider-paced class calibration: Calibrate the small choice set: The team presents labelled YES, NO, PAUSE and HELP targets with rest breaks and fits only the approved decoder.
- Read-back authorship check: Send and confirm messages: Alex selects a target; the partner reads it back through the established channel and Alex confirms or rejects before anyone acts.
- Eye-code comparison: Compare with best alternative: Measure confirmed accuracy, rate, correction time, fatigue, setup burden and privacy beside the eye-coded or board method.
Repeat plan
The clinical/research team controls calibration count, rest, dose and follow-up. Repeat only on its schedule; re-baseline after health, electrode/interface, decoder, vocabulary, supporter or environment changes.
Progress when
- Goal and consent are confirmed independently of the BCI.
- Configuration and responsible staff are logged before acquisition.
- During declared no-command trials, unintended output remains within the provider’s predeclared limit.
- Person-confirmed accuracy and correction meet the agreed threshold, no-command errors stay within the provider limit, fatigue is acceptable, privacy is maintained and fallback remains reliable.
Do not progress when
- Do not continue while this error remains: Treat decoder output as the person’s intention automatically.
- Pause until this correction works: Progress only on the provider’s criterion and Alex’s choice.
- This Brain–computer communication stop ends the block: Stop for pain, headache, skin change, seizure-like activity, distress, marked fatigue, new neurological symptom or participant withdrawal.
8 · Check the result
Measure what changed
Person-confirmed message accuracy, rate, correction, calibration/support time, home uptime, privacy and safe fallback versus best alternative access
How: Configured fixture: In a supervised session, Alex uses the approved BCI to choose YES, NO, PAUSE or HELP. Alex’s established eye-coded yes/no and alphabet board remain available. The provider logs “Check the complete system”, every “Send and confirm messages” result, the “Compare with best alternative” response and Person-confirmed message accuracy, rate, correction, calibration/support time, home uptime, privacy and safe fallback versus best alternative access. Log each BCI selection with decoder state, rejected artefact, partner read-back, fallback use, fatigue and setup time; count a message as correct only after Alex confirms it through the established channel.
Good result: Person-confirmed accuracy and correction meet the agreed threshold, no-command errors stay within the provider limit, fatigue is acceptable, privacy is maintained and fallback remains reliable.
This does not prove: Boundary for Brain–computer communication: “Person-confirmed message accuracy, rate, correction, calibration/support time, home uptime, privacy and safe fallback versus best alternative access” describes only In a supervised session, Alex uses the approved BCI to choose YES, NO, PAUSE or HELP. Alex’s established eye-coded yes/no and alphabet board remain available. It cannot establish “A brain implant reads thoughts perfectly”.
Self-check
- Without the example, demonstrate: Goal and consent are confirmed independently of the BCI.
- Find the fault in this attempt: “Treat decoder output as the person’s intention automatically.” Apply “Use read-back plus accept/reject before action.”; what changes?
- What evidence in the completed record shows that this is wrong: “Add many commands after one accurate block.”?
- Brain–computer communication stop decision: Stop for pain, headache, skin change, seizure-like activity, distress, marked fatigue, new neurological symptom or participant withdrawal.
9 · Stop, adapt or get help
Keep the safety boundary practical
Stop and get help
- Stop for pain, headache, skin change, seizure-like activity, distress, marked fatigue, new neurological symptom or participant withdrawal.
- Stop for repeated unintended output, unstable baseline, artefact, decoder drift, cybersecurity/account concern or unavailable fallback.
- Do not change implanted hardware, stimulation, decoder, vocabulary or session dose outside the responsible team.
Accessibility and adaptations
- Use the participant’s preferred conventional AAC for consent and confirmation and adapt display, dwell time, stimulus modality and rest through the team.
- Reduce choices or use partner-paced blocks without removing the participant’s right to pause, reject or end.
10 · Evidence and limits
Why these instructions are here
- primary research
Registered support for Brain–computer communication: “A high-performance speech neuroprosthesis”. It bears on Person-confirmed message accuracy, rate, correction, calibration/support time, home uptime, privacy and safe fallback versus best alternative access inside the Brain–computer communication fixture. It does not validate “A brain implant reads thoughts perfectly”.
A high-performance speech neuroprosthesis - primary research
Constraint for Brain–computer communication, drawn from “Long-term independent use of an intracortical brain–computer interface for speech and cursor control”: Earlier evidence was single-participant laboratory proof, and even long-term home use still required implantation, care-partner setup and adaptive software.
Long-term independent use of an intracortical brain–computer interface for speech and cursor control - official guidance
Privacy design for Brain–computer communication: minimise approved data in “In a supervised session, Alex uses the approved BCI to choose YES, NO, PAUSE or HELP. Alex’s established eye-coded yes/no and alphabet board remain available.” Keep Brain–computer communication provenance and access visible before interpreting Person-confirmed message accuracy, rate, correction, calibration/support time, home uptime, privacy and safe fallback versus best alternative access.
NIST Privacy Framework: A Tool for Improving Privacy Through Enterprise Risk Management, Version 1.0
Limits
- Brain–computer communication boundary: interpret “Person-confirmed message accuracy, rate, correction, calibration/support time, home uptime, privacy and safe fallback versus best alternative access” only for In a supervised session, Alex uses the approved BCI to choose YES, NO, PAUSE or HELP. Alex’s established eye-coded yes/no and alphabet board remain available.
- A successful result does not establish “A brain implant reads thoughts perfectly”.
- Brain–computer communication limiting finding: Earlier evidence was single-participant laboratory proof, and even long-term home use still required implantation, care-partner setup and adaptive software.
- No perfect-performance claim for Brain–computer communication: the evidence register does not make “Person-confirmed message accuracy, rate, correction, calibration/support time, home uptime, privacy and safe fallback versus best alternative access” universal, consequence-free or flawless in In a supervised session, Alex uses the approved BCI to choose YES, NO, PAUSE or HELP. Alex’s established eye-coded yes/no and alphabet board remain available.
- Scope remains Brain–computer communication: In a supervised session, Alex uses the approved BCI to choose YES, NO, PAUSE or HELP. Alex’s established eye-coded yes/no and alphabet board remain available. Recheck the comparator, support and “Person-confirmed message accuracy, rate, correction, calibration/support time, home uptime, privacy and safe fallback versus best alternative access” after any configuration change.
Open the complete canonical research register
- Primary empirical supportLimiting / contraryA high-performance speech neuroprosthesis
Francis R. Willett; Erin M. Kunz; Chaofei Fan; Donald T. Avansino; Guy H. Wilson; Eun Young Choi; Foram Kamdar; Matthew F. Glasser; Leigh R. Hochberg; Shaul Druckmann; Krishna V. Shenoy; Jaimie M. Henderson · 2023 · Primary research
- Primary empirical supportLimiting / contraryLong-term independent use of an intracortical brain–computer interface for speech and cursor control
Nicholas S. Card; Tyler Singer-Clark; Hamza Peracha; Carrina Iacobacci; Xianda Hou; Maitreyee Wairagkar; Zachery Fogg; Elena C. Offenberg; Leigh R. Hochberg; Sergey D. Stavisky; David M. Brandman · 2026 · Primary research
- Limiting / contraryOfficial boundary contextNIST Privacy Framework: A Tool for Improving Privacy Through Enterprise Risk Management, Version 1.0
National Institute of Standards and Technology · 2020 · Official standard
- Limiting / contraryOfficial boundary contextCybersecurity in Medical Devices: Quality Management System Considerations and Content of Premarket Submissions
United States Food and Drug Administration · 2026 · Official guidance
- Limiting / contraryOfficial boundary contextImplanted Brain-Computer Interface (BCI) Devices for Patients with Paralysis or Amputation — Non-clinical Testing and Clinical Considerations
United States Food and Drug Administration · 2021 · Official guidance
- Limiting / contraryOfficial boundary contextRecommendation on the Ethics of Neurotechnology
United Nations Educational, Scientific and Cultural Organization · 2025 · Official normative instrument
Read the complete evidence interpretation on the Power dossier.
Tutorial delivery controls
Learn, adapt, troubleshoot and resume
Progress is saved only in this browser on this device.
Step-by-step learner mode
Each activity includes its success check, a nearby accessible alternative and an “I’m stuck” correction path. Alternatives preserve the target where possible; when they change the task, Titan labels them as related rather than equivalent.
Set a person-chosen communication goal
Alex confirms the four-message task, acceptable effort, privacy boundary and preferred fallback through the established channel.
Goal and consent are confirmed independently of the BCI.
Goal and consent are confirmed independently of the BCI.
I’m stuck on this step
Reset: Re-read this authored instruction — “Alex confirms the four-message task, acceptable effort, privacy boundary and preferred fallback through the established channel.” — and its success check, then attempt only this step.
Possible snag: The result from “Alex confirms the four-message task, acceptable effort, privacy boundary and preferred fallback through the established channel.” does not yet meet this declared check: Goal and consent are confirmed independently of the BCI.
Correction: Return to the start of “Set a person-chosen communication goal”, reduce complexity or pace, and repeat only the part needed to satisfy: “Goal and consent are confirmed independently of the BCI.”
Stop / get help: Stop for pain, headache, skin change, seizure-like activity, distress, marked fatigue, new neurological symptom or participant withdrawal.
Check the complete system
The team verifies approved hardware, sensors or implant interface, decoder version, account, display, positioning, skin or clinical status and emergency procedure.
Configuration and responsible staff are logged before acquisition.
Configuration and responsible staff are logged before acquisition.
I’m stuck on this step
Reset: Re-read this authored instruction — “The team verifies approved hardware, sensors or implant interface, decoder version, account, display, positioning, skin or clinical status and emergency procedure.” — and its success check, then attempt only this step.
Possible snag: The result from “The team verifies approved hardware, sensors or implant interface, decoder version, account, display, positioning, skin or clinical status and emergency procedure.” does not yet meet this declared check: Configuration and responsible staff are logged before acquisition.
Correction: Return to the start of “Check the complete system”, reduce complexity or pace, and repeat only the part needed to satisfy: “Configuration and responsible staff are logged before acquisition.”
Stop / get help: Stop for pain, headache, skin change, seizure-like activity, distress, marked fatigue, new neurological symptom or participant withdrawal.
Record a no-command state
The provider captures rest or no-selection data and checks noise, artefact and unintended output.
A no-command baseline reveals unintended selections before any message is treated as the person’s communication.
During declared no-command trials, unintended output remains within the provider’s predeclared limit.
I’m stuck on this step
Reset: Re-read this authored instruction — “The provider captures rest or no-selection data and checks noise, artefact and unintended output.” — and its success check, then attempt only this step.
Possible snag: The result from “The provider captures rest or no-selection data and checks noise, artefact and unintended output.” does not yet meet this declared check: During declared no-command trials, unintended output remains within the provider’s predeclared limit.
Correction: Return to the start of “Record a no-command state”, reduce complexity or pace, and repeat only the part needed to satisfy: “During declared no-command trials, unintended output remains within the provider’s predeclared limit.”
Stop / get help: Stop for pain, headache, skin change, seizure-like activity, distress, marked fatigue, new neurological symptom or participant withdrawal.
Calibrate the small choice set
The team presents labelled YES, NO, PAUSE and HELP targets with rest breaks and fits only the approved decoder.
Each class has enough valid provider-accepted trials and rejected artefacts are counted.
Each class has enough valid provider-accepted trials and rejected artefacts are counted.
I’m stuck on this step
Reset: Re-read this authored instruction — “The team presents labelled YES, NO, PAUSE and HELP targets with rest breaks and fits only the approved decoder.” — and its success check, then attempt only this step.
Possible snag: Add many commands after one accurate block.
Correction: Progress only on the provider’s criterion and Alex’s choice.
Stop / get help: Stop for pain, headache, skin change, seizure-like activity, distress, marked fatigue, new neurological symptom or participant withdrawal.
Send and confirm messages
Alex selects a target; the partner reads it back through the established channel and Alex confirms or rejects before anyone acts.
Message meaning is person-confirmed, not decoder-assumed.
Message meaning is person-confirmed, not decoder-assumed.
I’m stuck on this step
Reset: Re-read this authored instruction — “Alex selects a target; the partner reads it back through the established channel and Alex confirms or rejects before anyone acts.” — and its success check, then attempt only this step.
Possible snag: Treat decoder output as the person’s intention automatically.
Correction: Use read-back plus accept/reject before action.
Stop / get help: Stop for pain, headache, skin change, seizure-like activity, distress, marked fatigue, new neurological symptom or participant withdrawal.
Repair false selection
On an incorrect HELP output, Alex uses PAUSE or the conventional channel, the partner cancels the message and the error enters the log.
Wrong output produces no external action and authorship is restored.
Wrong output produces no external action and authorship is restored.
I’m stuck on this step
Reset: Re-read this authored instruction — “On an incorrect HELP output, Alex uses PAUSE or the conventional channel, the partner cancels the message and the error enters the log.” — and its success check, then attempt only this step.
Possible snag: The result from “On an incorrect HELP output, Alex uses PAUSE or the conventional channel, the partner cancels the message and the error enters the log.” does not yet meet this declared check: Wrong output produces no external action and authorship is restored.
Correction: Return to the start of “Repair false selection”, reduce complexity or pace, and repeat only the part needed to satisfy: “Wrong output produces no external action and authorship is restored.”
Stop / get help: Stop for pain, headache, skin change, seizure-like activity, distress, marked fatigue, new neurological symptom or participant withdrawal.
Compare with best alternative
Measure confirmed accuracy, rate, correction time, fatigue, setup burden and privacy beside the eye-coded or board method.
The BCI is not judged by raw speed alone.
The BCI is not judged by raw speed alone.
I’m stuck on this step
Reset: Re-read this authored instruction — “Measure confirmed accuracy, rate, correction time, fatigue, setup burden and privacy beside the eye-coded or board method.” — and its success check, then attempt only this step.
Possible snag: Count only messages per minute.
Correction: Score accuracy, correction, fatigue and support together.
Possible snag: Remove it to prove BCI independence.
Correction: Practise immediate switch to the best alternative.
Stop / get help: Stop for pain, headache, skin change, seizure-like activity, distress, marked fatigue, new neurological symptom or participant withdrawal.
Finish with fallback and review
The provider stages decoder unavailability; Alex communicates one message through fallback, then the team checks symptoms and makes the next decision.
Fallback works and progression remains team- and participant-controlled.
Fallback works and progression remains team- and participant-controlled.
I’m stuck on this step
Reset: Re-read this authored instruction — “The provider stages decoder unavailability; Alex communicates one message through fallback, then the team checks symptoms and makes the next decision.” — and its success check, then attempt only this step.
Possible snag: The result from “The provider stages decoder unavailability; Alex communicates one message through fallback, then the team checks symptoms and makes the next decision.” does not yet meet this declared check: Fallback works and progression remains team- and participant-controlled.
Correction: Return to the start of “Finish with fallback and review”, reduce complexity or pace, and repeat only the part needed to satisfy: “Fallback works and progression remains team- and participant-controlled.”
Stop / get help: Stop for pain, headache, skin change, seizure-like activity, distress, marked fatigue, new neurological symptom or participant withdrawal.
Correct versus incorrect execution
These accessible process diagrams are built from the tutorial’s own right/wrong teaching. They are not anatomical illustrations and do not add technique beyond the canonical tutorial.
Confirm every consequential message with the person’s reliable channel.
Treat decoder output as the person’s intention automatically.
Begin with a small labelled vocabulary.
Add many commands after one accurate block.
Schedule rest and record setup/support time.
Count only messages per minute during the four-message BCI session.
Keep the established communication system ready.
Remove it to prove BCI independence.
Method-structure checklist
10 of 10 structural checks present
- Ordered, Power-specific instructions — present
- Every activity has a success check — present
- Materials or supplied records are declared — present
- Measurement or assessment rule is present — present
- Tutorial-specific troubleshooting is present — present
- Stopping or escalation boundary is present — present
- Every activity has an adjacent alternative — present
- Correct-versus-incorrect comparison is present — present
- Evidence context is bound to the Power record — present
- Planning metadata is present — present
The method-readiness band and presence checklist assess tutorial presentation and are separate from evidence quality for the underlying Power. They are automated editorial aids, not human approval.
Manual editorial sign-off: Pending. This tutorial must not display a human-approved state until an identified editor signs the exact content hash.