Embedded Files
– The individual and the fitness landscape
Abstract
In the century and a half since the idea of biological evolution was given a credible mechanism via which to operate, evolutionary thinking is now being applied to medically-related phenomena. However, the intellectual climate in which these ideas - in the form of evolutionary (Darwinian) medicine - operate is a complex one. The prevailing biological paradigm focuses strongly on the biological characteristics of populations, whereas that of medicine is primarily - although not exclusively - focused on the specific needs of the individual. These two approaches are quite different and some resolution of the dichotomy is necessary before evolutionary ideas can be fully integrated into medicine. What is missing - and where human biology may be able to help in the aforementioned resolution - is the realisation that it is from the characteristics of individuals that the characteristics of populations are derived and that it is via the individual that selection pressures operate.
This paper will consider, from a Darwinian biomedical perspective, the association between the individual and the population to which it belongs. In particular, a new two-dimensional biomedical model - which seeks to improve upon previous one-dimensional models - will be described and preliminary work considering its extension into a third dimension presented. The two-dimensional biomedical model, developed in recent years, seeks to characterise the individual as both a physical and sentient entity in evolutionary terms; its extension to a third dimension seeks to relate the individual to the selection pressures present in the environment in which that individual lives and to which it is variably exposed.
Introduction
In the century and a half since the publication of On the Origin of Species, the extent to which evolutionary ideas are being applied to biomedicine is still limited. The efforts of the proponents of evolutionary (or Darwinian) medicine notwithstanding, there appears to be greater interest displayed by the academic community – especially human biologists and anthropologists – in applying evolutionary ideas to biomedical topics (which they then make their own) than is displayed by the medical profession. What may be lacking perhaps are conceptual models which form a bridge between the different disciplines and which allow each to see human beings as the other sees them.
However, it is evident that the core model to which the clinical professions tend to hold is somewhat flawed. This model is the one-dimensional model of disease and health generally referred to as 'The Biomedical Model' and has been presented in different ways by different authors. Christopher Boorse, for example, uses a version of it while seeking to improve upon definitions of disease and health.
(From Boorse (1987))
Perhaps, as a heuristic – that is, a simple rule of thumb – this model offers some mental impression that aids clinical practice. However, as accurate biological models which offer critical ways of understanding more deeply the state of the individual, the current biomedical model is of very limited value.
But, having criticised this aspect of clinical thinking, I should not let evolutionary medicine go completely without criticism for one sees much the same models and ideas adhered to, albeit tacitly, in evolutionary medicine too; it has merely continued an adherence to the received wisdom. Evolutionary medicine, even with its unique insights has not, as yet, provided an alternative definition of disease or health. This is a potentially serious omission. If, as Randolph Nesse has proposed, evolutionary (or as it was then, Darwinian) medicine is "the enterprise of trying to find evolutionary explanations for vulnerabilities to disease", then we need an idea of what disease really is in order to fulfil this agenda.
Brief Description of a 2D Model
Over the past few years, a new biomedical model has been developed and a preliminary sketch has recently been published. Instead of the one-dimensional approach mentioned earlier, a two-dimensional plane has been proposed which depicts two key inter-related aspects of what it is to be an individual human being. The two dimensions in question are the physical state of the individual – depicted along the horizontal axis – and what can be referred to as the experiential state of that individual along the vertical axis.
We are sure to be familiar with the ideas encapsulated in the horizontal axis: it depicts how the organism is, physically/physiologically. However, this time, as one moves from left to right along this axis, the level of physical/physiological disorganization within the organism increases to a point at which, for that individual, life is increasingly less viable and a point ultimately reached when the individual dies.
The vertical axis depicts the experiential state of the individual: how the organism is known to itself at conscious and sub-conscious levels; it is an axis of self-reference. As one moves up this axis, the level of experiential stress increases to a point where consciousness is lost and the ability to respond behaviourally to the demands placed upon the individual ceases.
This notion of self-awareness is something that has been, perhaps, rather under-represented in biology while medicine has very much relied upon it. Without the ability to be self-aware – certainly in the form of feeling ill – one would never be aware that one needed medical attention. Self-awareness – or some other form of physiological self-reference – allows organisms, in their various ways, to monitor and respond to changing internal and external states and circumstances.
As well as having two axes instead of the usual one, this model reverses a common practice: that of equating positive attributes with the positive ends of an axis. Instead, as one moves along each axis, there is an increase in physical and experiential dis-organization – increasing levels of organismal disorganization (or entropy) are reached.
The intention is to depict what might be called the overall state of the individual, not as a point along a single line but as a changeable location on the plane within which the individual exists. The following example may help. An individual who feels well and has no physical problems belongs to the lower left of the plane [a]. (We might label this a 'healthy' state but how and why we label states as such is, in fact, another matter.) An individual who feels ill and who has a physical cause for this belongs to the upper right of the plane [b]. Then there are two other states which often – under the old model – caused clinicians problems: the state when the individual feels ill but for which there is no obvious physical cause [c] and the state when the individual feels well but has a lesion of some sort [d].
This model accommodates only the individual and while this was quite deliberate at the time of this model's formulation, a means of relating the individual to a wider context and to other individuals is the natural extension of the thinking.
Adaptive Landscapes
In the early 1930s, the American biologist Sewall Wright (1889-1988) – one of the architects of the neo-Darwinian synthesis – proposed the notion of the adaptive landscape. This was primarily a theoretical tool – another heuristic, this time in evolutionary biology – aimed at describing how different alleles can become more frequent within a population. As such, it was a means of illustrating natural selection in action.
With two axes representing two different genotypes, different interactions would lead, Wright suggested, to more or less (in his words) 'harmonious' combinations in given circumstances. Thus, the map represents areas (or peaks) of adaptation (denoted by a '+' sign) and other areas (troughs and valleys) of poor or non-adaptation (denoted by a '-' sign).
Wright's original formulation – although he seems to have, at times, been inconsistent with his use of this model – saw individual genotypes represented by single points plotted on the landscape while populations were represented as clouds of points typically found on or around an adaptive peak.
The adaptive landscape has been described as probably the most influential heuristic in evolutionary biology. It certainly has an abstract/conceptual appeal as a pictorial way of considering organismal characteristics under evolutionary pressures.
However, one, not uncommonly these days, encounters the term 'fitness landscape' rather than 'adaptive landscape'.
Fitness
'Fitness' has come to be used as evolutionary shorthand for 'reproductive success' and can be measured numerically as such. When referring to a fitness landscape, there is an assumption that different genotypes have clearly defined effects on reproduction. This, however, is a questionable assumption. While an inter-relationship between genotype and fitness does exist – as alluded to in this figure taken from Stearns and Hoekstra's (2005) Evolution: an introduction – the inter-relationship is affected through an individual's phenotype. Given what 'reproductive success' actually entails in practice, there is clearly no simple relationship between any of the three.
When Herbert Spencer (1820-1903) spoke, in 1864, of the 'survival of the fittest', he meant survival of those suited to their conditions – he was not thinking primarily of reproductive success. If one lacks fitness in this Spencerian (and later Darwinian) sense of 'being suited to one's habitat', reproduction is likely to be impeded. In this respect, fitness may be better understood in relation to phenotype rather than in relation to genotype.
Phenotypic Landscapes
In 1944, George Gaylord Simpson (1902-1984) did, indeed, propose a notional landscape based upon phenotypic characters. Graphically, the landscapes have much the same appearance except that the determinants of the contours are quite different. While Wright's adaptive landscape – based upon genotypes – was theoretically acceptable, Simpson's use of phenotypic traits is conceptually more satisfying. Associations between physically discernable features and reproductive success are more obvious. Unfortunately, this aspect of Simpson's work has been overlooked somewhat and has only recently been revived.
It is worth remembering here the modern definition of a phenotype as being any observable characteristic or trait of an organism. This includes anatomical, physiological and behavioural properties and, as such, must include the states represented by the two-dimensional biomedical model described earlier. It is at the level of the phenotype that illness, disease and health occur. Thus, if phenotype has a bearing on fitness, illness, disease and health as aspects of the phenotype must also be included in any complete understanding of fitness.
Combining The Different Models
I have described a new two-dimensional biomedical model and I have described the use of (metaphorical) landscapes as ways of representing groups of individuals in terms of their biological fitness. We can bring the two together but, in order to do so, we must first address one somewhat tacit feature of these landscapes as they have so far been used.
Both the adaptive and phenotypic landscapes think in terms of a static organism and also, one might reasonably argue, an idealised one – not idealised in the sense of being perfected but idealised in the sense of the organism being seen at its theoretical optimal state. An old, sick and dying individual, using Wright's original version of the adaptive landscape, can be represented, genetically, at an adaptive peak but while, in evolutionary terms, one can see Wright's point – that the individual's genotype is well adapted – at the same time the same individual, in strictly phenotypic terms cannot be deemed to be at any sort of peak or optimal state. Indeed, it is sub-optimally that most, if not all, organisms actually live out their entire existence. Thus, one may be genetically adapted to a given environment but still fail to be completely suited to it due to pathology, injury or some other mishap. Organisms do not inhabit a single fixed point on a fitness landscape. They exist at different sub-optimal levels; within a patch on the landscape. Later in life, they descend – via perhaps any number of routes – into one or other trough or valley, to a position of being no longer suited to their habitat. It is these real life states that the two-dimensional biomedical model seeks to describe.
This, then, is an association of two models from two different but increasingly inter-linked areas of human biology: the biomedical and the evolutionary. What has been sought is a conceptual association between the two: the two-dimensional biomedical model provides a way of conceptualising the state of the individual which can then be overlaid on a fitness landscape. The two models have not been linked, however, in the sense of their various axes being joined.
With the two-dimensional model in mind, an individual who feels well and has no physical problems [a] and an individual who feels ill with due physical [b] cause may be represented accordingly on the fitness landscape.
However, the clinically problematic situations mentioned earlier – that of the state when the individual is ill but with no obvious physical basis for that illness and the state when the individual feels well but has a lesion of some sort – can also be considered in relation to the notion of a fitness landscape with perhaps more interesting results. In the first scenario (when there is illness but no obvious physical cause [c]), the fitness of the individual is compromised whereas in the second case (when the individual is feeling well but also has a lesion [d]), fitness is not compromised – at least not for the time being. Here, there is a clear contrast between what is important clinically and what is important biologically – that is, evolutionarily. An asymptomatic lesion can cause medical concern yet not necessarily impede reproductive potential whereas an illness of little clinical interest may temporarily exclude one from reproductive activity.
Given the reliance on received wisdom when it comes to our notions of disease and health – it is important to realise such differences: that medical values and biological realities are not necessarily the same and that an unqualified exchange of such between the two disciplines is not strictly viable.
Conclusion
The idea of associating the new two-dimensional biomedical model with that of the fitness landscape represents fairly preliminary work in that this is its first outing. Much has been omitted and there is much still to explore. However, I am confident that the general notion is correct – or at least that it can serve as a useful heuristic linking biomedicine and evolutionary biology conceptually.
Given the time in which it was written, one of the hallmarks of Darwin's Origin of Species was its emphasis on individual examples. This is not surprising given that Darwin worked in an era before population-based biology took centre stage. Given the new ways in which biology and medicine are beginning to intersect, it may be timely to reconsider the place of the individual in the wider scheme of things, especially if ideas of illness, disease and health – which only properly relate to individuals – are to be understood in proper perspective.
Bibliography
Boorse, C. (1997). A Rebuttal on Health. In J. Humber & K. Almeder (Eds.), What is Disease? (pp. 3-134). Totowa, New Jersey: Humana Press.
Simpson, G. (1944). Tempo and Mode in Evolution. New York: Columbia University Press.
Stearns, S., & Hoekstra, R. (2005). Evolution: an introduction (2nd ed.). Oxford: Oxford University Press.
Wright, S. (1932). The roles of mutation, inbreeding, crossbreeding and selection in evolution. Proc. 6th Int. Congr. Genet., 1, 356-366.
Wright, S. (1988). Surfaces of Selective Value Revisited. American Naturalist, 131, 115-123.
Page updated
Google Sites
Report abuse