Physical versus Computational Complementarity

The dichotomy between endophysical/intrinsic and exophysical/extrinsic perception concerns the question of how a model—mathematical, logical, computational—universe is perceived from inside or from outside, [71, 65, 66, 59, 60, 68, 67]. This distinction goes back in time at least to Archimedes, reported to have asked for a point outside the world from which one could move the earth. An exophysical perception is realized when the system is laid out and the experimenter peeps at the relevant features without changing the system. The information flows on a one-way road: from the system to the experimenter. An endophysical perception can be realized when the experimenter is part of the system under observation. In such a case one has a two-way informational flow; measurements and entities measured are interchangeable and any attempt to distinguish between them ends up as a convention. The general conception dominating the sciences is that the physical universe is perceivable only from inside. This view reduces the exophysical perception to a theoretical illusion. The more plausible perception, i.e. an endophysical one, suffers from a “self-referential” disease as any intrinsic measurement causes uncontrolled, and maybe uncontrollable, “disturbances” to the entity intended to be measured. This paper, the first in a proposed series, discusses some limitations and trade-offs between endophysical/intrinsic and exophysical/extrinsic perceptions, in both physical and computational contexts. We are building our work on Moore “gedanken” experiments [50] in which the universe is modeled by a finite deterministic automaton. A new type of computational complementarity, which mimics the state of quantum entanglement, is introduced and contrasted with Moore’s computational complementarity. Computer simulations of both types of computational complementarity are developed for four-states Moore automata.