CONNECTOME LAB / EXPERIMENT 01

FlyTris.

A tiny player at the controls.

Learning observed in every replicate

256 neurons · 9089 connections
8 learned output weights

READY TO PLAYDrag to look around
01 /THE WIRED PLAYERSix legs. Six connections. One game.
Leg taps illustrate the saved rotation, column and drop.A little head shake for every cleared row.
HUMANS & ONE VERY SMALL OPPONENT

The scorebook.

Three-minute ranked matches. Score = lines cleared; pieces placed break ties.

Loading the scorebook…

Human vs. fly

Humans 0 lines

Fly 0 lines

No completed matches yet.

The fly’s brain stays the same at every pace. Faster play allows more decisions before time runs out.

Chart totals include every recorded match in the selected window and difficulty.

Easy800 ms gravity · fly every 5s

Player / dateHumanFly

Medium500 ms gravity · fly every 3s

Player / dateHumanFly

Hard250 ms gravity · fly every 1.5s

Player / dateHumanFly

Top 10 human scores per difficulty, paired with the fly’s score from the same match. Dates use your local time. Names are player-provided.

THE MEASUREMENTS

Learning, measured.

RunBeforeAfterGain [95% CI]Training
110.1240.71+40.58 [33.33, 48.62]1.1 min
227.46103.83+96.38 [86.75, 103.88]1.4 min
330.0868.58+68.50 [58.33, 78.50]1.3 min

Mean lines cleared on identical, unseen piece sequences. This replay is the first held-out game from the first training run.

42.25Trained raw-feature control
0.04Random placements
UNDER THE SHELL

What actually learns?

Board features enter a fixed fly-derived spiking circuit. Eight learned output weights choose a placement from its activity.

The fly animation illustrates recorded decisions. It is not a biomechanical simulation. Removing recurrent connections did not consistently hurt performance in these trials, so fly wiring has not been shown to be essential.