where it stands is what it holds.
it doesn’t trade.
it walks.
A reconstructed Drosophila nervous system is placed in a room of tokenized stocks and left to do what it evolved to do.
tick 146 690 · heading 86.1° · turn −1.8° · DNp01 quiet · walking
The market, as a place.
Five tokenized stocks, five rings on fixed lines down the room. A ring’s diameter is its volume and its depth is its momentum, so a stock the market is pushing comes forward at you and a quiet one falls back toward the door. Ink coverage is its weight in the book. The fly walks the floor between them, and its distance to each ring is the allocation.
- NVDA11.3%1.35M
- TSLA15.0%2.68M
- AAPL30.1%896K
- AMZN29.0%1.08M
- MSFT14.5%1.02M
position -0.210, 0.007 · five weights summing to one
How it walks.
Three things have to be true for any of this to mean anything. Each of them is a decision we made, not a fact we found.
- 01
The room is lit by volume.
The fly is never handed a price. Each tick the room is rendered from where it is standing and which way it is facing, and that picture is all it gets. A ring that is being traded heavily is wide and densely inked, so it takes up more of the visual field and pulls harder on the circuitry that steers. A ring nobody is trading is small, thin and far away. Volume does not instruct the fly. It changes what there is to see.
- 02
Heading comes from EPG cells.
A population of neurons in the ellipsoid body holds a single bump of activity on a ring, and where that bump sits is the fly’s heading. Turning the body moves the bump; seeing a landmark pins it. We do not read the heading out and put it back — the bump persists because of the recurrent wiring already present in the connectome. PFL cells compare that heading against a goal and push harder on one side than the other, and the difference is a turn.
- 03
Position is the portfolio.
Where the fly stands on the floor is converted to weights by inverse squared distance to the five rings, normalised to sum to one. Nothing else decides the book. Two things follow, and neither is incidental: standing still trades nothing, because unchanged coordinates produce unchanged weights and no swap is submitted; and every rebalance is a walk, because a body cannot cross the room instantly. The turn rate of the fly is the rate limit on the portfolio.
Recent ticks.
Spike counts, heading, position and weights are committed in the same transaction that moves the book. Every row links to it.
- 146 69086.1°−1.8°AAPL30.1%0xd6f4d97b…57aaef
- 146 68987.8°−1.1°AAPL23.6%0xd54cbdc9…16b5de
- 146 68889.0°+1.2°AAPL17.8%0x018f2998…eb8128
- 146 68787.8°+8.3°AAPL13.3%0x70be1da5…45e19d
- 146 68679.5°+4.0°AAPL10.5%0x64a8aadf…b34a19
- 146 68575.6°−2.6°AAPL8.6%0xbbf4390a…bd0388
- 146 68478.2°−8.2°AMZN56.0%0x8fe5dd0d…e7839e
- 146 68386.4°−4.0°AMZN55.2%0x5bb74803…41362b
- 146 68290.4°−2.1°AMZN51.4%0x9978f434…a602d5
- 146 68192.5°−9.3°AMZN47.5%0xcaf2b3da…895f89
- 146 680101.7°+177.2°AMZN45.1%0xa19061a0…7258ae
- 146 679284.5°−16.4°MSFT38.0%0xecf4516a…5a1def
Where this comes from.
- MaleCNS v1.0 connectome
- Complete brain and ventral nerve cord of an adult male Drosophila melanogaster. Over 166,000 neurons, roughly 125 million synaptic connections; the largest brain map by neuron count. Publicly released 8 June 2026. Licensed CC-BY.
- The paper
- “Sexual dimorphism in the complete connectome of the Drosophila male central nervous system”, Cell, 3 September 2026.
- Who built it
- FlyEM at HHMI Janelia Research Campus, University of Cambridge Department of Zoology, MRC Laboratory of Molecular Biology, and Google Research. Reconstructed from electron microscopy, proofread by human experts, with AI assembling millions of 2D images into 3D.
- How we read it
- Annotations and connectivity queried through the neuPrint API against dataset male-cns:v1.0. Neuron identifiers shown in the ledger are the released identifiers, unchanged.
A connectome is anatomy, not physiology. It records which neurons touch which others and how often; it does not record what any of them do. Sensory encoding, membrane dynamics and the mapping from body position to portfolio weight are engineering choices, and any of them could be wrong.
No profitable strategy has been demonstrated and none is claimed.