What it does
- Assumption deconstruction
- Falsifiable hypothesis design
- Steelman counterarguments
- Cross-disciplinary synthesis
NeurOnstein
Alzheimer's Disease Reasoning
Parametric-knowledge-only hypothesis engine for Alzheimer's disease — challenge dominant assumptions, then propose falsifiable mechanisms from molecular to evolutionary scales.
Who it's forNeurodegeneration Researchers · Translational Scientists · Drug Discovery Teams
Built for
- AD researchers
- Translational neuroscientists
- Drug discovery teams
- Academic hypothesis labs
Value
- No literature retrieval
- Non-amyloid-centric reframes
- Experiment-ready predictions
- Confidence-calibrated output
How work changes
Same desk. Different first pass.
Traditional
Every page, every identifier, by hand.
Out
Literature review
Out
Manual hypothesis drafting
Out
Internal lab debate
Out
Design validation study
Out
Run experiments
Out
Iterate on findings
NeurOnstein
The model takes the first pass. People take the exceptions.
Out
Pose the biological question
Out
NeurOnstein deconstructs hidden assumptions
Out
Generates falsifiable hypotheses
Out
Steelmans strongest counterarguments
Out
Proposes realistic experimental tests
Out
Rates confidence level
Summarizing amyloid-centric narrativesInventing testable mechanisms worth experimentally challenging
Used for
Late-onset AD
Reframe vascular-metabolic, immune, and sleep hypotheses beyond default amyloid cascade framing.
Resilience anomalies
Mechanistic hypotheses for amyloid-positive elders who remain cognitively intact.
Multi-scale synthesis
Connect molecular, cellular, systems, metabolic, environmental, and evolutionary layers in one falsifiable claim.