Banner
    The Problem Of Kin Selection Theory
    By Gerhard Adam | June 12th 2009 05:57 PM | 12 comments | Print | E-mail | Track Comments
    About Gerhard

    I'm not big on writing things about myself so a friend on this site (Brian Taylor) opted to put a few sentences together: Hopefully I'll be able...

    View Gerhard's Profile
    Kin selection is one of those special considerations derived from “selfish gene theory” that postulates that it is the degree of relatedness between organisms that will determine the likelihood that altruistic actions will occur. This also clearly implies the existence of a social group, of some type, so it isn’t expected that it would play a role between members of different species, or among asocial animals (although it could).

    In general the idea of “kin selection” is that individuals are more apt to behave altruistically to “blood relatives” than to others in the interest of propagating their genes into future generations. A classic example occurs in eusocial insects where sterile females help maintain the colony for the reproductive queen.

    However the use of eusocial insects is problematic since there is no possibility of an unrelated insect ever being present in the colony(1). Therefore there is no way to determine whether the behavior is due to simple proximity, kin relatedness, or response to the social group itself. In effect, there is no way to test kin selection theory in this case since every insect is, by definition, kin.

    The nature of kin selection theory is described by Hamilton’s rule:

    C < R x B

    The values in this expression are interpreted such that C represents the cost in fitness to the actor and B the benefit in fitness for the recipient, while R represents the degree of genetic relatedness between the two. In this case, fitness is reproductive capacity.

    Without getting into all of the calculations and terms of this relationship, one of the biggest questions focuses around the variable R. While this is defined as being the degree of genetic relatedness, there is nothing in this equation to require it, nor is it derived by any other means.

    In other words, R could just as easily be used to describe the longevity of a relationship between individuals (kin or not), or it could be used to describe the ratio of brown-eyed to blue-eyed individuals. In all these cases, a solution could be derived, but it wouldn’t necessarily have any basis for extrapolation into the real world.

    The most obvious problem with kin selection theory is that it requires that an animal be capable of assessing the degree of “relatedness” to an individual before acting. In the case of a strong familial group, this may not be difficult to determine, however the difficulty occurs when it becomes virtually impossible to distinguish an action as being based on “kin” versus the social group to which an individual belongs. This seems to have been shown in a recent study (2) where familiarity resulted in more positive interactions than kinship did, which coincides with most intuitive expectations.

    There is little doubt that many social groups may have initially formed as families and extended families, so that there would be a high likelihood that there is a genetic relationship amongst a group’s members. This does not have universal application and consequently many of the members of a group may have no genetic relationship to each other. In this case their behavior must be construed as being oriented towards the survival of the group (or at least close individuals), rather than any notion of “kin”.

    An additional difficulty occurs because in the quest for future mates, other social groups may be the source of such mates. As a result, the concept of “kin” must extend beyond the immediate genetic relationship to include others with whom future genetic exchange may occur.

    Regarding the altruistic actions themselves, part of the confusion comes from considering altruistic behaviors as being anomalous and trying to rationalize their existence by postulating a kin relationship as evolutionary justification. In point of fact, many altruistic behaviors exist to ensure the survival of the group, because without the group there may be no individual chance of survival. Therefore, it may be quite reasonable for an individual to perform an act which is altruistic to ensure group survival, since their own survival would be jeopardized without the group anyway. It is also important that to recognize that costs or benefits are only potentially realized and cannot be assessed until after the action is taken.

    By considering the potential cost of an altruistic act, we tend to overlook the fact that a “successful” altruistic act may convey great status to the individual and be of significant benefit. However this is rarely considered because it is presumed that altruistic actions that may have ulterior motives are somehow less altruistic or that they are only applicable to zero-sum circumstances.

    In addition, motives or intent can play no role in assessing an action. Simply because we can never truly know what is behind another organisms actions. We can speculate or make assumptions, but it is impossible to know with any degree of confidence what the actual motivating factor is. The difficulty comes from having to infer a particular “state of mind” which can never be absolutely determined.

    While it is relatively easy to see how “kin selection” can factor into animals that rear their offspring, it is a much more difficult assessment when evaluating adult behaviors. Related or not, how would one distinguish an act as benefiting a particular individual over the group to which they belong? It is obvious that all actions are largely going to be between individuals, so to suggest that there is such a specific reason for it, goes beyond our ability to assess it accurately.

    Most importantly, there can be no action undertaken based on the certainty of a particular outcome. The altruist cannot know the consequences of their actions any more than that the benefit to the recipient can be assured. Therefore, we must conclude that whatever factors are at work, they must be less ambitious in explaining the cooperative nature of these species. While there is no question that genes are a fundamental element in natural selection, it may be that by placing undue emphasis on this aspect of evolution we are simply being too reductionist in our views.


    (1) Notable exceptions related to aphids or other symbiotic relationship. However, these represent special cases much like domesticated animals to humans.

    (2) Kin recognition versus familiarity in a solitary mustelid, the European polecat Mustela putorius

    Comments

    Hank
    You would have enjoyed EO Wilson at the 'enigma of altruism' panel at the World Science Festival last evening.  Another panelist, even giving an overview of kin selection, got a jolly, "It's just that it's the wrong theory" muttered at him.
    Want more no-nonsense, independent science? Buy Science Left Behind
    logicman
    In general the idea of “kin selection” is that individuals are more apt to behave altruistically to “blood relatives” than to others in the interest of propagating their genes into future generations.
    Gerhard: as you know, I am working on a related idea: the evolution of language. 

    Part of my research is directed towards the role played by mirror neurons.  If A exhibits a behaviour, and B copies that behaviour to B's overall advantage, altruism and selfishness  are neither necessary nor sufficient, nor even relevant, to explain the act of copying.

    Primatologists have frequently observed primates copying simple human behaviours.  With no desire on the part of a human to teach, and no evidence of any 'desire' on the part of a primate to learn, learning takes place.  There is no altruism, no selfishness.  The mimicry 'just happens'.

    Humans are excellent mimics and fast learners.  A child sees someone burn their hand on a hot surface and instantly loses all desire to experiment.  This is accidental demonstration and incidental learning.  The event witnessed may even be a fatal accident.  There is obvious detriment to the 'teacher' and obvious evolutionary advantage to the 'student'.  But it would be madness to assign values of altruism and selfishness to the participants.

     Sometimes, all we can say about the acquisition of survival value by an individual or group at a cost to another individual or group is: stercus accidit.
    Steve Davis
    Insightful. Incisive. Incendiary. This theory that was alleged to explain the spread of altruism, worse still, the ONLY way altruism can spread, is now blown out of the water. And it was the "scientific" basis of selfish gene theory. Well done.
    Steve Davis
    Another panelist, even giving an overview of kin selection, got a jolly, "It's just that it's the wrong theory" muttered at him.
    That's a great story Hank, but Ed should not be muttering it, he should be thundering it! 
    I feel that you mix altruistic behavior where you eventually get a benefit and purely altruistic behavior.
    Kin selection is successful at explaining pure altruism. That is altruism in which you cannot receive benefit - for example because you knowingly sacrifice your life. You cannot explain such behavior by any eventual benefit to yourself.

    Bye,

    Arent

    Gerhard Adam
    Kin selection is successful at explaining pure altruism. That is altruism in which you cannot receive benefit - for example because you knowingly sacrifice your life. You cannot explain such behavior by any eventual benefit to yourself.
    Not true.  There are numerous instances of sacrifice that have nothing to do with kin (i.e. fire/police departments, soldiers in combat, etc.).  Each has to do with supporting the group with which they are associated over themselves.  It is the group that has meaning (i.e. military unit or police partners, etc.).

    The point is that such altruism exists, regardless of whether there's a satisfactory answer to explain it.  However, it is clear that "kin selection" cannot account for these instances, and quite frankly it is far more likely that a stranger dies saving a loved one (i.e. fire/police/rescue) than a family member.

    In addition, the notion of a benefit (except tangentially as in reciprocal altruism) is overstated.  In many cases, the existence of the group becomes more important than the individual because without the group there can be no individuals.  Therefore, purely self-interested or selfish behavior would invariably lead to extinction.  In addition, future mates come from strangers that are not genetically related, so there must be a sense of cooperation between unrelated members simply to ensure enough diversity of mating partners exist.  Kin selection (in its strictest form) would fail because it could only guarantee in-breeding partners.

    Regarding benefits, consider that of the 10^13 cells in your body that cooperate to produce a healthy person that such a state is only threatened when cells begin acting "selfishly".  Under normal circumstances cells willingly die and even commit "suicide" (Programmed Cell Death) to preserve the larger group. 

    The most obvious problem with kin selection theory is that it requires that an animal be capable of assessing the degree of “relatedness” to an individual before acting.

    No it doesn't. Evolution is often anthropomorphized like this by people who don't understand it, but that's not the way the theory actually works. It describes tendencies to do things for individuals who are probably related. Wolves don't do a genetic analysis of each other to calculate how much of a risk they should take for the group. Organisms don't even need to be conscious for kin selection to work. They just have to evolve to do risky things in situations where there is a probability that it will lead to a benefit to their genes.

    http://books.google.com/books?id=bb86AAAAIAAJ&pg=PA63

    Under normal circumstances cells willingly die and even commit "suicide" (Programmed Cell Death) to preserve the larger group.

    Yes. That's the essence of selfish gene theory. All the cells in the body have exactly the same genes, and thus have evolved to easily sacrifice themselves so that their genes can survive in other cells. Individual cells became colonies, which became multicellular organisms like us, for exactly this reason.

    You seem to have a fundamental misunderstanding of what selfish gene theory claims. Have you even read anything on the subject?

    Gerhard Adam
    You seem to have a fundamental misunderstanding of what selfish gene theory claims. Have you even read anything on the subject?
    Selfish gene theory is, at best, a novel perspective, but it certainly has no basis in evolutionary processes.  In the first place, even Dawkins admits that his definition of genes is so vague and nebulous that virtually anything he says could be deemed correct. 
    "I said that I preferred to think of the gene as the fundamental unit of natural selection, and therefore the fundamental unit of self-interest.  What I have now done is to define the gene in such a way that I cannot really help being right!"
    (Dawkins, The Selfish Gene, pg 33)
    However, the primary problem is that there is no constant definition of a gene, nor is there a predictable method of assessing what is expressed.  Therefore a particular sequence of nucleotides does not represent any particular trait and consequently it is difficult to argue that there is any basis for assuming that it is "selfish" with respect to ensuring it's propagation into future generations. 
    All the cells in the body have exactly the same genes, and thus have evolved to easily sacrifice themselves so that their genes can survivein other cells.
    Which is exactly why selfish gene theory is absurd.  You're arguing that your basis for selfish genes is that they all cooperate and willingly sacrifice themselves so that there is a possibility that some of the genes in the germ line get propagated.  You cannot claim selfishness while requiring all manner of cooperation to achieve your stated objective.  That's simply absurd.
    Evolution is often anthropomorphized like this by people who don't understand it, but that's not the way the theory actually works.
    There's no anthropomorphizing taking place here.  Kin selection requires the ability to recognize kin, otherwise the null hypothesis says that normal probabilities could also explain such a phenomenon and doesn't require postulating a "kin selection" criteria.  Your use of probabilities clearly suggests the null hypothesis and refutes the concept of kin selection, since you're suggesting that probabilities are the determining factor whether or not the animals are related.  If kin cannot be differentiated from others due to proximity, then there is no basis for suggesting that kin carries a privileged position with respect to selection or cooperation.

    Considering Maynard Smith's examples:
    One of the clearest examples of kin selection occurs in a bacterial plasmid (Maynard Smith );
    The problem with this is that it doesn't describe many of the cooperative or social situations where this also has to apply.  In other words, one of the flaws in kin selection theory is that the survival of an organism isn't based solely on itself, but rather on the survival and existence of a sufficiently large number of others for sexual selection and reproduction.  Therefore, kin selection is counter-productive towards survival, since it is the survival of the group which will determine whether or not extinction is a potential threat.  Maynard Smith is using an example of asexual reproduction which begs the question.  In addition, bacterial reproduction has an extremely high probability that adjacent cells are related, therefore kin selection isn't necessary, but proximity will suffice.
    "there is no possibility of an unrelated insect ever being present in the colony"

    well exactly. any non related insect who tries to invade is killed. not exactly altruistic.

    Gerhard Adam
    What are you talking about?  The point of "kin selection" is to argue that such members would altruistically sacrifice themselves in defense of each other (in the extreme) because they are related.  If the population you're examining only consists of related individuals, then any conclusions are suspect since you haven't examined any alternatives.

    In fact, your statement regarding non-related insects is wrong when examining Argentinian Fire Ants that owe a great deal of their success to supporting multiple queens when it suits them.  This clearly suggests that kinship is insufficient to account for the cohesion of the nest.  There is no question that there is some mechanism that allows identification of individuals associated with a particular nest and that this provides cohesion.  However, there is no evidence to suggest that it is explicitly genetic.
    "the population you're examining only consists of related individuals"

    the REASON it consists of only relatives is because they kill all non relatives.

    multiple queens are still relatives

    Gerhard Adam
    That's a circular argument.  You cannot reason that kin selection is the biological method for sacrifice or altruism and then only examine an organism that is related.
    multiple queens are still relatives
    That's just wrong.

    Multiple-queen (polygyne) colonies of the introduced fire ant Solenopsis invicta present a paradox for kin selection theory. Egg-laying queens within these societies are, on average, unrelated to one another, and the numbers of queens per colony are high, so that workers appear to raise new sexuals that are no more closely related to them than are random individuals in the population.  
    http://www.springerlink.com/content/gkxu7xrq7w8njmmc/
    The point all along has been that we already know that there appears to be a higher degree of cooperation among kin, but the important question is whether this is a true interpretation based on genetics, or whether its applicable simply because of the degree of familiarity.  In other words, using a very simple example, if genetically unrelated animals are placed together to be reared, will they behave as if they were kin to each other?  If yes, then obviously, genetics is not a determining factor in such behavior and simple proximity will do.  It should be clear that a significant number of animals will tend to remain geographically close to each other resulting in a much higher incidence of relatedness in future encounters.  So, once again, the question then is whether relatedness is favored or simply coincidental because of geographic distributions.

    An obvious example can be observed in humans where we can consider whether a parent devotes any more or less resources to a child that is genetically related, adopted and genetically unrelated, or genetically unrelated (but the parent isn't aware of it).  The point is that if there is no substantive way for these three behaviors to be differentiated, then it is hard to argue that kin selection operates in any meaningful way.