Locked-In Syndrome Assistive Communication

Locked-in syndrome assistive communication is the use case in EP 6: Data Science & AI Talk where brain-signal models might eventually help people who can think but cannot express themselves because their bodies are paralyzed. Paulina Nemkova presents this as the original motivation behind her team’s EEG Brain Reading work.

The concept extends Assistive AI by moving from voice or wearable accessibility into brain-signal interpretation. The source keeps the current capability modest: object-category classification is a step toward communication support, not proof that AI can read complete thoughts.

Key Claims

• The user’s agency is the center of the use case: the model matters only if it helps expression.
• Brain-signal classification has a clearer ethical grounding when tied to communication barriers rather than surveillance or spectacle.
• The current technical result should be treated as early-stage because EEG categories are narrower than ordinary language or thought.
• Assistive applications still need AI Verification and Human-Driven Scientific AI because false or overconfident interpretation could harm vulnerable users.

Connections

• EEG Brain Reading and Paulina Nemkova - source research and source voice.
• Assistive AI - broader accessibility frame.
• AI For Science, AI Verification, and Human Judgment Under AI - scientific and human-responsibility context.
• Research Replication Integrity - evidence boundary for brain-signal claims.