Phenotypic red herrings, natural selection and the biological bases of behaviour: a cautionary cry.

I have spent much of this week reading about the various schools of thought that have appeared, grown, merged or divided over the last sixty or seventy years on the topic of the biological or genetic bases or, put more succinctly, components of behaviour. After sorting through the important, and I must say from my own point of view admirable (in some places more than others of course), work of people like Lorenz, Tinbergen, Wilson, Hamilton, Maynard-Smith, Dawkins, the list goes on, and importantly those that opposed their work, a few consistent areas of concern became readily apparent to me. Some on the side of the ‘sociobiologists’ (an out of fashion term that I feel fits nicely in the modern context never the less) and others on the side of those that criticise their work as reductionist or misanthropic when applied to humans (social scientists and the like).

The general issue is as follows. I will explain how it affects both sides of this argument shortly.

As a post-graduate in the field of biological anthropology with a firm background in biology, and as a consequence the rudiments of genetics, I’m acutely aware of pleiotropy, epistasis, allometry and other factors that can see a given gene influencing multiple phenotypic traits, genes influencing one another and morphological traits affecting each other out of mechanical necessity. In short, I’m cautious about viewing a given trait (physical or behavioural should someone believe it to have some non-environmental basis) and exclaiming ‘this trait was selected for by natural selection and must therefore have a survival function’, this survival function fallacy births innumerable evolutionary scenarios that are likely to be absolute garbage. When wrangling with evolutionary ideas for a given organism I find it useful to have one of two approaches.

1) This is the most desirable approach though it is very rarely practical. Work from the genes up. Gene x produces protein y that influences cells of type a, b and c and thus tissue type z and so on and so forth. From this perspective one can see where the genetic influence begins to wane and the environmental factors begin to have more influence. Think of a connective tissue disorder as a convenient example, like Marfan syndrome, where a problem gene causes defective fibrillin produciton. This single gene, coding for a single protein, has multiple phenotypic effects that are then acted upon by the environment. I stress here that this same principle applies to genetic influencers of behaviour, a gene relating to a neural structure that has survival benefits would also drag with it numerous other effects that can be neutral or even slightly negative providing the cost doesn’t outweigh the benefit. Taking a magnifying glass to one of the incidental traits (what Gould would have called a ‘spandrel’) and looking for an evolutionary scenario that fits its selection is participating in the survival function fallacy.

2) Don’t start with the organism, start with the environment. If you cannot work from the genes up, work from the ecosystem down. Before conjuring evolutionary explanations for traits in a given organism, first make sure its environment is understood. This may help to sort important traits from those that are simply coincidental (pleiotropic perhaps etc.). Choose virtually any book on human evolution and you will see the consequences of starting the analysis with the organism. Scientists who start with any given hominin’s remains and then try to impose environments onto these remains will have a hard time sorting the wheat from the chaff, often with laughable results that damage reputations and waste the time of researchers and readers alike.

Gould & Lewontin summed up the perils of an ‘adaptionist’ approach very nicely in their 1979 paper ‘The spandrels of San Marco and the Panglossian paradigm: a critique of the adaptionist programme’.

So, how does all of this relate to the work of those looking for the biological or genetic bases of behaviour? Put simply, are we really sure that any given behavioural pattern or what the ethologists would have called ‘instinct’ (convenient language, this is an outmoded notion) HAS a survival function? Was it simply pulled along with something that did? Is it a spandrel? This is perhaps my greatest criticism of a field of work that I do, otherwise, have great faith in. Workers in modern sociobiology and its offspring could perhaps benefit from working from one of the two approaches I outlined above.

As for the opponents of sociobiology of any modern type under whatever monicker it now takes (behavioural ecology, evolutionary psychology, gene-culture co-evolution etc.), I find a slightly different criticism for their approach which also relates to an understanding of genetics. Firstly, as any good geneticist or biologist of current times will stress, no modern day researcher in these fields truly believes in the false dichotomy of nature or nature. It is agreed upon that there are biological components and environmental components, the question is a matter of scale. Secondly, something I see time and time again in the criticisms of biological approaches to understanding behaviour is this old chestnut ‘the gene to phenotype to selection model of natural selection when applied to behaviour ignores choice, culture and other human attributes’. As mentioned above, this view of genetics is at its most innocent a gross simplification and at its worst absolutely erroneous. Unfortunately it is also widespread on both sides of the fence. Being erroneous, it is of course, not the basis of a good argument for or against anything. I also will reiterate here that no good biologist or biological anthropologist denies the role of culture and choice in shaping human beings, even with the numerous models for the survival function of culture and ideas about comparative cognition. Suffice it to say, meme models and comparative cognitive studies, if carried our responsibly, do not have to damage our concept of free will or our pride as a species.

Time will tell how things progress in this field, and I will be watching closely. I will provide a further exploration of this idea next week with an example from real world research, the work of Belyaev on the domestication of silver foxes.


~ by confusedious on August 5, 2011.

2 Responses to “Phenotypic red herrings, natural selection and the biological bases of behaviour: a cautionary cry.”

  1. Isn’t Darwin’s concept that mutation and evolution are successful when they allow an organism a better fit with it’s environment still a good overarching philosophy? Maybe we need to clarify ‘environment’ to be more than nature, but to include the social, psychological, ecological, etc surroundings of an organism.

    I think that notion of ‘best fit’ is the fulcrum and maybe where paleoanthropologists get themselves into trouble. How can one determine ‘best fit’ if one doesn’t truly know the environment beyond what all the paleo-sciences tell us it is? Given a rich set of paleo-factors (geology, climate, botony, and more), I’m sure different scientists will develop different habitats, usually based on their field of expertise. How can we ever really know?

    • What you say of Darwin’s concept is true, but it is often horribly misapplied by people who ignore the realities of genetics. It is true that Darwin’s ideas are at the foundation of our modern understanding of evolution, that does not mean however that ‘the new synthesis’ (where we stand after the work of Dobzhansky, Haldane, Mayr, Huxley etc.) relies entirely on the basic notions of natural selection. A decent understanding of genetics leads one to realise that pleiotropy, for example, produces a vast array of spandrels that need not have an adaptive function, they need only be associated with a trait that does have one.

      You are quite right about the definition of environment, the truth is that modern evolutionary biologists and anthropologists always mean a combination of their ‘physical’ environment (i.e. predators, food sources, climate and the like) and their social and psychological environment.

      Your last point is the real kicker. We can’t ever really know, all we can do is try to be good scientists and work within the limits of what we have. That is why I suggest always starting with reconstructions of the physical environment (pollen, geology and the like as suggested) and then working down. From here, if paleo-scientists are able to find multiple individuals (a big ask in most cases, finding one fossil hominid is amazing enough) they can compare the individuals with the context of the physical environment in mind, differences of sexual dimorphism for example allow one to make some fairly accurate assumptions about interactions between the sexes. Strontium isotope analysis of teeth can be used as well to understand the distribution of males and females should more than one individual (or at least their teeth) be found. It is then much more appropriate to start thinking about physical traits of the individuals and whether or not they were subject to heavy selection pressure. In particular, I think it would be useful to have some knowledge of how pleitropy affects humans and at least apes also and to always keep allometry in mind.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: