Neurons And Intelligence: A Birdbrained Perspective
Read the original on Slate Star Codex →
Follow-up reading
↳ [PARTIALLY RETRACTED] Cortical Neuron Number Matches Intuitive Perceptions Of Moral Value Across Animals — extends the neuron-count finding · Mar 2019
Summary
A clear explainer (off an AI Impacts report + Herculano-Houzel's neuron-counting work) on what makes animals intelligent. Not brain size or encephalization quotient but CORTICAL NEURON COUNT: birds pack tiny dense neurons (up to 200k/mg vs human 25k vs elephant 7k), so a crow rivals a monkey, and blinded raters judge birds as smart as similarly-neuroned primates -- so the mammalian-cortex vs avian-pallium structural difference barely matters. A cortical-neuron-count ranking matches intuitive animal-intelligence orderings; Dicke & Roth add packing density and conduction velocity (IPC). The pilot-whale anomaly (2x human cortical neurons) is the loose end. AI implication: no need to mimic cortical columns, and AI ability may scale ~linearly with compute.
Why this score
Quality 73 · Strong. A clear, genuinely interesting, well-synthesized science explainer with a crisp takeaway (cortical neuron count) and a forward-looking AI-scaling hook. Strong.
Claude’s paradigm shift 50 · Moderate. Moderate. Reports others' findings, but the synthesis and the AI-scaling framing are a fresh angle.
Real-world impact 2 · Minor. A clear, well-synthesized science explainer with a crisp takeaway (cortical neuron count, not brain size, tracks intelligence — birds rival primates) and a forward-looking AI-scaling hook. Conceptual/pedagogical influence within rationalist discourse, no material change — low RWI.