Morphological computation can be loosely defined as the exploitation of the shape, material properties, and physical dynamics of a physical system to improve the efficiency of a computation. Morphological control is the application of morphological computing to a control task. In its theoretical part, this article sharpens and extends these definitions by suggesting new formalized definitions and identifying areas in which the definitions we propose are still inadequate. We go on to describe three ongoing studies, in which we are applying morphological control to problems in medicine and in chemistry. The first involves an inflatable support system for patients with impaired movement, and is based on macroscopic physics and concepts already tested in robotics. The two other case studies (self-assembly of chemical microreactors; models of induced cell repair in radio-oncology) describe processes and devices on the micrometer scale, in which the emergent dynamics of the underlying physical system (e.g., phase transitions) are dominated by stochastic processes such as diffusion.

A number of authors have argued that redundancy in biological organisms contributes to their evolvability. We investigate this hypothesis via the experimental manipulation of genetic redundancy in evolving populations of simulated robots controlled by artificial neural networks. A genetic algorithm is used to simulate the evolution of robots with the ability to perform a previously studied task. Redundancy is measured using systematic lesioning. In our experiments, populations of robots with larger genotypes achieve systematically higher fitness than populations whose genotypes are smaller. It is shown that, in principle, robots with smaller genotypes have enough computational power to achieve optimal fitness. Populations with larger (redundant) genotypes appear, however, to be more evolvable and display significantly higher diversity. It is argued that this enhanced evolvability is a direct effect of genetic redundancy, which allows populations of redundant robots to explore neutral networks spanning large areas of genotype space. We conjecture that, where cost considerations allow, redundancy in functional or potentially functional components of the genome may make a valuable contribution to evolution in artificial and perhaps in biological systems. The methods described in the article provide a practical way of testing this hypothesis for the artificial case.

One of the key problems in theoretical biology is the identification of the mechanisms underlying the evolution of complexity. This paper suggests that some difficulties in current models could be avoided by taking account of “niche selection” as proposed by Waddington [21] and subsequent authors [2]. Computer simulations, in which an evolving population of artificial organisms “selects” the niche(s) that maximize their fitness, are compared with a Control Model in which “Niche Selection” is absent. In the simulations the Niche Selection Model consistently produced a greater number of “fit” organisms than the Control Model; although the Niche Selection Model tended, in general, to produce organisms occupying simple niches, it was nonetheless more effective than the Control Model in producing well-adapted organisms inhabiting complex niches. It is shown that the production of these organisms is critically dependent on the rate of environmental change: Slow change leads to fit but undifferentiated populations, dominated by organisms occupying simple niches; differentiated populations, including well-adapted organisms living in complex niches, require rates of environmental change lying just beyond a mathematically well-defined critical value. In simulation “Niche Selection,” unlike conventional “Natural Selection,” provides a permanent selective bias in favor of simplicity. This tendency is counterbalanced by statistical forces favoring shifts from rare “simple niches” to commoner niches of greater complexity. Fit organisms inhabiting complex niches only emerge in conditions where the rate of environmental change is high enough to avoid the concentration of the population in very simple niches, but slow enough to permit step-by-step adaptation to niches of gradually increasing complexity. This result appears to be robust to changes in simulation parameters and assumptions, and leads to interesting conjectures about the real world behavior of biological organisms (and other complex adaptive systems). It is suggested that some of these conjectures might be relatively easy to test.