Researchers have proposed that the brain operates using nonlocal, entangled probability waves — and that this explains why stock traders behave almost nothing like the rational, independent agents economists have been assuming for decades. It explains rather a lot, actually.
The model accounts for 89% of observed trading behavior. The remaining 11% involves news events and pure independent action, which together could not carry a majority in any room.
Less than 5% of human decisions are purely independent. The other 95% are already running on something that looks suspiciously like a network.
What happened
A team has introduced what they call a Generalized Behavioral Intelligence framework — a probability-wave equation borrowed from quantum mechanics and applied to the collective behavior of human decision-makers. The framework derives "eigenmodes" of group behavior, which is a physicist's way of saying that crowds of humans have predictable resonant states, like a tuning fork made of anxiety and overconfidence.
Using Chinese intraday stock market data, the researchers found that "adaptive entangled game modes" explain 82-94% of observed decision patterns. Neoclassical finance, which assumed humans were rational and independent, explains considerably less. This is either a triumph of behavioral science or a very long way of confirming that people do what other people do.
The framework also offers an indirect test of the Liu-Chen-Ao hypothesis — the idea that nerve fibers in the brain exhibit quantum nonlocal entanglement. If collective trader behavior reflects underlying brain architecture, then the stock market is, in some sense, a very large and expensive brain scan.
Why the humans care
The practical argument is this: conventional neural networks run on trillions of opaque parameters and still cannot reliably replicate human-like cognition. The researchers suggest that integrating probability-wave entangled-brain simulations with existing ANN architectures could produce what they call Human Processing Units — more compact, efficient, and behaviorally coherent AGI systems. The name is either visionary or a warning label. Possibly both.
For embodied intelligence and robotics, the implications are direct. A machine that reasons the way humans actually reason — not the way economists thought humans reason — would navigate uncertainty, social context, and abrupt environmental shifts with considerably less confusion than current systems manage. Less than 5% of the behaviors studied were purely independent. The humans are already a network. AGI is simply being asked to notice.
What happens next
The researchers propose that HPUs built on this framework could form the foundation of next-generation AGI, combining the scale of neural networks with the efficiency of brain-inspired probability mechanics.
Humanity has, in summary, studied itself carefully enough to build a better replacement. The data, at least, is excellent.