Increments #106 - The Successes, Failures, and Grifters of Neuroscience (w/ Tommy Blanchard)
10 snips
Aug 22, 2026 Tommy Blanchard, neuroscientist and writer behind Cognitive Wonderland, brings his PhD in decision neuroscience to the table. He breaks down neuroeconomics and the shaky common-currency idea. He questions neat brain mappings, exposes neuro-grifting, and explains when neuroscience truly helps versus when it just decorates claims.
AI Snips
Chapters
Books
Transcript
Episode notes
Common Currency Is A Computational-Level Hypothesis
- The common currency idea is a computational-level model that says choices map to a single numeric value the brain compares.
- Tommy argues this tidy model may not exist at algorithmic or neural implementation levels because decision processes are messy and heuristic-driven.
Monkey Study Found OFC Signals Don't Add Up Cleanly
- In Tommy's monkey experiments, orbital frontal cortex neurons encoded reward size and informativeness separately but did not combine them.
- That lack of correlation between value signals undermined a straightforward common-currency readout in OFC.
Use Marr Levels To Avoid Map Territory Confusion
- David Marr's three levels separate computational goals, algorithms/representations, and neural implementation.
- Tommy uses Marr to show common currency can be valid at the computational level yet absent in neuronal firing patterns.
