Biological Systems as Physical Data Processors: Evolution as a Case of Embodied Computing

Abstract

The prevailing view of biological evolution is not unlike bricolage/pastiche/tinkering—in sharp contrast with rational engineering. Yet, different paths often lead to solutions that coincide or converge whether they emerge from naturally-occurring evolution or rationally designed. Such a conjunction—often presented as an anecdote—reveals the ability of biological systems to explore solution spaces and gravitate toward information-rich attractors, which can be found through different routes. This scenario evokes one of embodied computing, a non-conventional type of data processing in which the solution to a problem is not delivered through numerical calculations but through its implementation in a material object. Once left to undergo a physical process the object manages a large number of parameters for reaching a multi objective optimum through thermodynamically-driven processes. The course of information is thus a physical flow and the outcome is a physical currency. The consequences of this notion for bioengineering are remarkable.