Synaptic bundle theory for spike-driven sensor-motor system: More than eight independent synaptic bundles collapse reward-STDP learning

Researchers developed a spike-driven sensor-motor system that identifies critical limits for neuronal learning. The study found that learning collapses when the number of motor neurons or independent synaptic bundles exceeds certain thresholds, providing insights into biological spike-based control mechanisms.

• →Learning systems fail when motor neurons or synaptic bundles exceed critical limits in spike-based control.
• →Fewer motor neurons increase learning failure probability but enable faster learning when successful.
• →The number of weight updates moving opposite to optimal direction quantitatively explains learning outcomes.
• →Spike-driven control systems face inherent learning collapse issues when applied to artificial actuators.
• →Identifying parameter ranges for functional spike-based learning could unlock previously inaccessible neural functions.