Structure alone supports efficient visual computation in the Drosophila visual system

2026-10-07

AI Summary

This study investigates the extent to which synaptic connectivity structure determines visual computation. By integrating the whole-brain connectome of Drosophila with an eye model, we construct a purely anatomy-driven computational framework in which only neuronal gain parameters are optimized. Multi-task visual processing performance is evaluated using linear decoders and synthetic data benchmarks. We demonstrate that, under wiring economy constraints, biologically precise connectivity yields significantly superior representational capacity compared to random networks, revealing that topological and geometric features jointly anchor efficient computational manifolds. Notably, without end-to-end training, the model successfully supports color, shape, and numerosity discrimination tasks. These findings validate that neural circuit architecture fundamentally underpins efficient visual computation.

Abstract

Understanding the extent to which measured synaptic wiring determines computation remains a central challenge. Here, we couple the proofread adult Drosophila melanogaster connectome to an anatomically faithful model of its eye. Visual information is inputted in the eye model, then passed to the connectome, and finally read from a Kenyon-cell-centered linear decoder. This creates a connectome-only model in which the anatomical graph and eye geometry are fixed and only scalar synaptic gains and neuronal thresholds may be learned. The model supports multitask vision, including color discrimination, shape classification, and numerical discrimination that follows a ratio-dependent scaling characteristic of approximate number perception. To test whether precise connectivity is consequential under wiring economy, we compare the biological graph to randomized ensembles that increasingly preserve biological synaptic constraints. At matched wiring cost, the biological network consistently yields higher accuracy, whereas less constrained rewiring surpasses it at the cost of inflated wiring. These findings indicate that the measured connectivity and eye geometry jointly set efficient operating points for visual computation.

Problem

• connectome
• visual computation
• wiring economy
• Drosophila
• synaptic connectivity

Innovation

• Connectome
• Drosophila visual system
• Wiring economy
• Multitask vision
• Anatomically faithful model