27 lines
2.4 KiB
Markdown
27 lines
2.4 KiB
Markdown
|
# Rate-Distortion Theory in Markovian Networks: Why Fitness Beats Truth
|
||
|
|
|
||
|
|
**Target Venue:** *Journal of Theoretical Biology*
|
||
|
|
|
||
|
|
## Abstract
|
||
|
|
Donald Hoffman's "Fitness Beats Truth" (FBT) theorem demonstrates that perceptual systems are tuned for survival fitness rather than veridical representations of objective reality. We provide a strict information-theoretic foundation for FBT using Shannon's Rate-Distortion Theory. By treating biological perception as an optimal lossy compression algorithm across a Markovian agent network, we mathematically prove that an agent minimizes its metabolic computational cost (the bit rate $R$) subject to a strict distortion constraint (survival probability $D$). Veridical perception requires an unbounded bit rate, exceeding biological ATP metabolic constraints. Thus, the non-veridical "desktop interface" is the unique optimal solution to the rate-distortion function in a competitive fitness landscape.
|
||
|
|
|
||
|
|
## 1. Introduction
|
||
|
|
Evolution selects for perceptual interfaces that hide complexity (Hoffman et al., 2015). While this is proven via game theory, the thermodynamic and computational constraints driving this selection must be formalized.
|
||
|
|
|
||
|
|
## 2. The Rate-Distortion Formulation
|
||
|
|
Let the objective network state be $X$ and the agent's internal representation be $Y$. The agent seeks to minimize the mutual information $I(X;Y)$ to conserve metabolic energy, subject to an expected distortion constraint $\mathbb{E}[d(X,Y)] \le D_{max}$, where $d(X,Y)$ is the fitness penalty of misrepresentation.
|
||
|
|
The rate-distortion function is:
|
||
|
|
$$
|
||
|
|
R(D) = \min_{p(y|x) : \mathbb{E}[d] \le D} I(X;Y)
|
||
|
|
$$
|
||
|
|
|
||
|
|
## 3. The Thermodynamic Cost of Truth
|
||
|
|
A veridical representation implies $D \to 0$, forcing $R(D) \to H(X)$ (the full entropy of the environment). According to Landauer's principle and the ATP costs of neural spike generation, supporting a bit rate $H(X)$ requires infinite metabolic energy. Consequently, $p(y|x)$ must be a highly lossy mapping (a homomorphism).
|
||
|
|
|
||
|
|
## 4. Conclusion
|
||
|
|
Fitness beats truth because truth is metabolically bankrupting. The perceptual interface is exactly the optimal probability channel $p(y|x)$ that solves the rate-distortion optimization problem for a biological organism.
|
||
|
|
|
||
|
|
## References
|
||
|
|
1. Hoffman, D. D., Singh, M., & Prakash, C. (2015). *The interface theory of perception*. Psychonomic Bulletin & Review.
|
||
|
|
2. Shannon, C. E. (1959). *Coding theorems for a discrete source with a fidelity criterion*. IRE National Convention Record.
|