Real neurons proved far messier than the logic-gate model predicted
Evidence
To Pitts’s disappointment, the neuroscience work of the 1950s, notably a landmark study of the optic nerve of the frog—done by none other than his best friend, Jerry Lettvin—will show that neurons appear to be much messier than the simple “true” or “false” circuits he envisioned. Perhaps propositional logic—its ands, ors, and nots—was not, ultimately, the language of the brain, or at least not in so straightforward a form. This kind of impurity saddened Pitts. — Brian Christian, The Alignment Problem, 16
Categorization
- XREF: Connects to the book’s central theme of alignment—the gap between idealized models and messy reality that recurs throughout AI history. — connects to: The brain’s all-or-nothing neuron firing resembles binary logic — Brian Christian, The Alignment Problem, 12; Any logical expression can be implemented by a neural network — Brian Christian, The Alignment Problem, 15; Prefrontal cortex neurons selectively represent only task-relevant information — Nowak, M. A., & Sigmund, K, Evolution of Indirect Reciprocity by…, 124
- SEED: Angle: the emotional cost to the theorists whose elegant models are falsified by empirical reality—a human dimension of scientific progress.
- corroborated: 2 independent passages support this claim — Brian Christian, The Alignment Problem, 16
Broader Connections
- The brain’s all-or-nothing neuron firing resembles binary logic
#contradicts(0.80) — Brian Christian, The Alignment Problem, 12 - Any logical expression can be implemented by a neural network
#qualifies(0.40) — Brian Christian, The Alignment Problem, 15