Heavy-tailed connections support robust neural communication

Heavy-tailed synaptic connectivity reduces communication errors under synaptic noise, with robustness determined by a single network statistic in minimal neuron models.

• Why it matters: Rare, much stronger synapses are ubiquitous, but it was unclear whether this heavy-tailed wiring actually improves neural communication when noise perturbs synaptic transmission.
• What they did: The authors combine analytical results and simulations in minimal single-neuron models, showing that noisy connections produce an error rate depending only on a “robustness” network statistic.
• The result: For a fixed number of incoming synapses, broader connection-strength distributions are linked to fewer errors; across several connectomes, real neurons are consistently more robust than comparable null networks, and in Drosophila melanogaster robustness increases with distance from the sensory periphery.