Information theory, predictability and the emergence of complex life.

Luís F Seoane,Ricard V Solé

doi:10.1098/rsos.172221

Abstract

Despite the obvious advantage of simple life forms capable of fast replication, different levels of cognitive complexity have been achieved by living systems in terms of their potential to cope with environmental uncertainty. Against the inevitable cost associated with detecting environmental cues and responding to them in adaptive ways, we conjecture that the potential for predicting the environment can overcome the expenses associated with maintaining costly, complex structures. We present a minimal formal model grounded in information theory and selection, in which successive generations of agents are mapped into transmitters and receivers of a coded message. Our agents are guessing machines and their capacity to deal with environments of different complexity defines the conditions to sustain more complex agents.

Highlights

Simple life forms dominate our biosphere [1] and define a lower bound of embodied, self-replicating systems
Despite the obvious advantage of simple life forms capable of fast replication, different levels of cognitive complexity have been achieved by living systems in terms of their potential to cope with environmental uncertainty
Our core finding is that the complexity of the guessers that can populate a given environment is determined by the complexity of the latter. (In information-theoretical terms, the complexity of the most efficiently replicated message follows from the predictability of the channel.) Back to the fast replication versus complexity question, we find environments for which simple guessers die off, but in which more complex life flourishes— offering a quantifiable model for real-life excursions in biological complexity

Summary

Introduction

Simple life forms dominate our biosphere [1] and define a lower bound of embodied, self-replicating systems. This curve trivially dictates the average survival or extinction of the optimal 1-guessers in infinite, unstructured environments as a function of the cost–reward ratio α ≡ c/r (grey area subtended by the solid diagonal line in figure 1c) Note that this parameter α encodes the severity of the environment— i.e. how much does a reward pay off given the investment needed to obtain it. Note that any more complex guessers (like the ones described in successive sections) would always fare worst in this case: they would potentially pay a larger cost to infer some structure where there is none This results in narrower survival areas qualitatively represented by shades of grey subtended by the discontinuous lines in figure 1c. Instead of developing specific models for each of these alternative implementations, we resort to mathematical abstractions based on bit-strings, whose conclusions will be general and apply broadly to any chosen strategy

Evolution and information theory

Results

Numerical limits of guesser complexity

Evolutionary drivers

Discussion