General Systems Theory

The central goal of most scientific disciplines is to understand systems, i.e. ensembles of interacting elements. Today, this statement sounds almost trivial, yet in biology at least, the importance of the systems concept has been established only relatively recently. A key figure was Ludwig von Bertalanffy, a biologist and philosopher, who wrote a series of seminal articles in the first half of the 20th century in which he argued that complex phenomena in biology (and indeed any other scientific field) invariably result from systems and could only be understood properly through a mathematical description of how system behavior emerged from the interactions of its constituent elements. Demonstrating the existence of system isomorphisms, i.e. general mathematical descriptions that explained the dynamic behavior of very different kind of systems at different scales and across fields as diverse as physics, biology, economy and sociology, he introduced a very general framework that became known as general system theory (see the collection of essays in von Bertalanffy 1969). By the 1940s, the systems concept had experienced a scientific breakthrough in biology and led to the rise of cybernetics, "the science of control and communication in the animal and the machine" (Wiener 1948; Ashby 1956).

Today, biology uses the systems concept to address questions at all levels, from the molecular level to whole organisms and populations. The systems concept is now so omnipresent in biology that a recent special issue of the journal Science on systems biology renewed von Bertalanffy's (1969) previous diagnosis: "The [systems] concept has pervaded all fields of science and penetrated into popular thinking, jargon, and mass media" (Chong & Ray 2002).

But what exactly is a "system" and why is the systems concept so useful for framing scientific questions? A general, yet informal, definition is that a system is a set of elements which interact with each other in a spatially and temporally specific fashion. Before we attempt a formal definition of a system in the next section, let us remind ourselves that one of the classic scientific methods is to "analyze" a given phenomenon, i.e. to break it down into atomic units and processes that can be investigated independently of each other. This approach is appealing because it reduces a complex problem to a set of simpler problems, each of which can be addressed under conditions which can be controlled more easily for potentially confounding influences. For example, if one wanted to understand the physiological properties of a single neuron, one might decide to isolate it from its environment (e.g. let it grow in a dish) and then map its responses to currents injected into various parts of its dendritic tree. Unfortunately, this analytic approach cannot fully predict the neuron's behavior when it is part of a neural system, e.g. in the brain, and thus interacts with other neurons. When part of a system, the response of an individual neuron to a particular synaptic input (or injected current) u\ depends on the spatial and temporal distribution of inputs ui . ..un that its dendritic tree receives from other neurons. If these additional inputs occur sufficiently close in time and space to ui , they will affect the magnitude of the postsynaptic potential elicited by ui , either linearly (by spatio-temporal summation) or nonlinearly (e.g. by changing the opening probability of voltage-gated channels) (Magee & Johnston 2005). In other words, the connectivity in the system mediates effects that cannot be predicted by studying a single neuron. Similar scenarios can be described for any other scientific field, for example biochemistry. Having studied a set of different biochemical processes in isolation, one would not necessarily be able to predict their collective dynamics. The problem is, as above, that different processes may interact, e.g. one process may change the substrate/product ratio of another process, or the efficacy of an enzyme that is relevant for a particular process may change due to the presence of allosteric (in)activators that are produced by a second process or due to dynamic changes in gene expression mediated by a third process.

In summary, the general problem of analytical procedures in science is that they are blind to predicting the consequences arising from interactions between the elements in a system. Analytical procedures therefore need to be complemented with a theoretical framework that takes into account both the connectivity between the elements and external perturbations in order to achieve a mechanistic explanation of the dynamics of the system as a whole. This framework is provided by general system theory.

Was this article helpful?

0 0

Post a comment