Energy budgets govern synaptic precision and its regulation during plasticity
New model shows synaptic reliability follows a precise σ⁻² ∝ E⁵ scaling law, governed by metabolic costs.
Deep Dive
Researchers James Malkin, Cian O'Donnell, and Conor Houghton developed an energy-constrained framework showing synaptic precision is fundamentally limited by metabolic budgets. Analyzing five plasticity datasets, they found synapses cluster near a minimal-energy boundary, with precision scaling as σ⁻² ∝ E⁵. This demonstrates plasticity systematically reallocates energy based on signal change, enabling prediction of post-plasticity variance from energy allocation alone.
Why It Matters
Provides a fundamental principle linking metabolism to neural computation, with implications for designing more efficient, brain-inspired AI systems.