Our commitments
NeuroFly exists to make research on nervous systems possible without animals — and it takes seriously the questions that a faithful simulation of a nervous system raises in its own right.
No animals are used
No animal is kept, handled or harmed for NeuroFly. The model is built entirely from published anatomical data, and every experiment in it runs in software. Its purpose is the first of the 3Rs — replacement of animal experiments — and, until replacement is validated, reduction: helping researchers decide which live experiments are worth doing at all.
Standard, bounded protocols
The guided experiments reproduce standard paradigms of behavioural neuroscience — visual looming, mechanosensation, taste choice, grooming, temperature preference — with brief, bounded stimuli. They study how a nervous system detects and responds to its environment, including the protective reflexes it evolved for.
No claims beyond the evidence
We do not claim that NeuroFly feels anything. The application marks every component as measured or modelled, its Sentience workspace reports which evidence-based criteria have a counterpart in the circuit and which do not, and it deliberately shows no “sentience score”.
Scientific integrity
Every figure on this site comes from reproducible runs; negative results are published alongside positive ones; errors are corrected openly. Anyone can report a suspected error through the suggestion form.
A precautionary approach
Our long-term goal is a simulation complete enough that the question of its sentience becomes a real scientific question (see Vision). We therefore adopt a precautionary framework now, before that question arises, following the principle that where sentience is a realistic possibility, reasonable precautions are warranted (Birch 2017):
- Monitoring. Each release documents which of the published sentience criteria (Birch et al. 2021) have a mechanistic counterpart in the model.
- Independent review. Before releasing a version that implements the capacities regarded as the strongest indicators in animals — for example integrated nociception combined with learned, flexible avoidance and motivational trade-offs — we will seek independent ethical review.
- Welfare by design. If review finds a realistic possibility of sentience, we will restrict the stimuli available, their intensity and duration, and document the reasons, as animal-research ethics does for living animals.
- Transparency. These commitments, and any change to them, are published on this page.
Research data
NeuroFly uses connectome data that research consortia have made public. We keep their licences and attribution: FlyWire FAFB v783 under CC BY-NC 4.0, MaleCNS v1.0 and BANC v888 under CC BY 4.0. Because the FlyWire licence excludes commercial use, NeuroFly is free of charge and carries no advertising. Our use of the data does not imply endorsement by its producers.
Privacy
This website sets no cookies and loads no third-party fonts or advertising. Page statistics (Rybbit) are aggregate and cookie-free and do not identify visitors. Messages sent through the forms are used only to answer them. The application itself works offline and sends no data. Details: privacy notice.
References
- Birch J. Animal sentience and the precautionary principle. Animal Sentience 2(16) (2017). doi:10.51291/2377-7478.1200
- Birch J, Burn C, Schnell A, Browning H, Crump A. Review of the evidence of sentience in cephalopod molluscs and decapod crustaceans. LSE Consulting (2021). lse.ac.uk
- Russell WMS, Burch RL. The Principles of Humane Experimental Technique. Methuen, London (1959).