Defining Fitness

There is a pretty interesting discussion going on in Noyau regarding the many definitions of “fitness” in evolutionary biology. It would be a shame for it to be lost in that particular venue here at TSZ. At the risk of being censored by the admins for posting too many OPs in one month I thought I’d start this thread.

Here’s my take so far:

Allan Miller was charged by phoodoo with resorting to different definitions of fitness. Allan denied the charge and when asked for a definition of fitness Allan provided one. Allan later stated that his definition only properly applied to asexual species.

Others chimed in to say that the definition of fitness depends on the context, which hardly seems to contradict what phoodoo was saying.

My own position is that fitness has its definition within a particular mathematical framework. My position is also that fitness can be defined generically but that such a definition is tautological. Special definitions of fitness are required to make the concept testable.

Here’s hoping we can move the discussion about fitness out of Noyau.

93 thoughts on “Defining Fitness

  1. walto,

    Francis the poodle at t1 is not identical to any of his offspring. He’s a reproducing entity and (roughly) the more successfully reproducing offspring he has, the higher HIS fitness. If you’re interested in the fitness of his whole line, you look at the reproductivity of additional generations.

    On what basis do you judge his fitness prior to his reproductive success assuming he eventually generates many new poodles. Can you demonstrate correlation between reproductive success of his mother and father translates to his reproductive success?

    If he contains his mother and fathers genes and some random genetic variation what is the chance that this variation improves his fitness? What is the chance it degrades his fitness?

    If we cannot predict fitness then what are we really talking about?

  2. colewd,

    If we cannot predict fitness then what are we really talking about?

    I don’t know why prediction has much to do with it. Things can be causal without being precisely predictable. Why do an experiment? Why not just ‘predict’?

  3. walto: Francis the poodle at t1 is not identical to any of his offspring. He’s a reproducing entity and (roughly) the more successfully reproducing offspring he has, the higher HIS fitness.If you’re interested in the fitness of his whole line, you look at the reproductivity of additional generations.

    So yeah, YOU have to be clear.And it’s really not that complicated.You’re hunting for obscurities where there actually aren’t any.

    Oh my goodness, apparently it IS that complicated, because you still aren’t getting it.

    If Francis the poodle is the entity, then Francis the poodle is NOT increasing at all. Instead we are mixing Francis the poodle, with max the bulldog, then max and Francis, with Rover and Belle, and then Rover and Belle and Francis and Max, and mixed with ten other dogs, and not ONE of them is Francis the poodle!

    And if we stretch this out, you can see why I made fun of the idea that we are all the shrew from the Jurassic. If everything after Francis the poodle, is still Francis the poodles fitness, then everything after the shrew is still the fitness of the shrew. So every time your cousin has a baby, the shrew is becoming more fit.

    That’s the implication of a silly definition.

  4. You’re just being intentionally obtuse, phoodoo. Again, no entity is or is claimed by anybody to be identical to its offspring. A common name (like ‘dog’ ) may apply to both, but that’s not a problem for anybody but you.

  5. walto:
    You’re just being intentionally obtuse, phoodoo. Again, no entity is or is claimed by anybody to be identical to its offspring.A common name (like ‘dog’ ) may apply to both, but that’s not a problem for anybody but you.

    I think you are intentionally being obtuse, how about that?

    If I am testing the fitness of some shrew from the Jurassic, do I count EVERYTHING that has come since then?

    If I am testing the fitness of Francis the poodle, when do I start and when do I stop? If I am looking at the fitness of Australopithecus, do I have to count everything from Lucy until now?

    You are claiming that we should just say dogs and be done with it. Or just shrews and be done with it, and then once they are no longer shrews, we stop giving them credit?

    When did the shrew stop being a shrew?

    Behold, the fittest human on Earth:

    https://www.livescience.com/26929-mama-first-ancestor-placental-mammals.html

  6. Did anyone read the paper I uploaded? It’s a pretty old paper by now, but a citation classic (>500 citations according to Web of Science). No comments?

    Oh well. Creationists like Mung and Phoodoo (haha Phd = voodoo) are typically too innumerate to understand technical scientific papers, but of course that doesn’t stop them from criticizing that which they don’t understand. I wish I had the patience to keep up with that.

  7. Ido Pen,

    Well, perhaps it would be helpful if you said what it is exactly in the paper that you wanted to call people’s attention to. Do you even know?

    Just posting a link to something, and saying, here, go read this, I think its good, is not really something that should be on everyone’s homework assignment.

    Go read David Berlinski, I think its useful.

  8. phoodoo:
    If I am testing the fitness of some shrew from the Jurassic, do I count EVERYTHING that has come since then?

    For about the seventh time, it depends what question you’re trying to answer. If you want to know the fitness of that entire line, I’d think you might need an estimate of that number of descendents. So what? If you think that’s a useless calculation and want to know the fitness of a subset only, then no. Different universes of discourse, different answers. How many more times would you like to be told that?

  9. phoodoo: Well, perhaps it would be helpful if you said what it is exactly in the paper that you wanted to call people’s attention to. Do you even know?

    Yes, he knows. The paper discusses an approach in which you establish whether a resident population with a particular life history strategy (and what is that?) can be invaded by a novel mutant. It’s fantastic work, but definitely a little too involved for this crowd (not only the anti-evolutionists).

  10. Ido Pen,

    phoodoo is right. I love the idea of the ESS, but the paper definitely needs some elaboration since it is way too technical for all but a few specialists here on the forum.

  11. colewd: On what basis do you judge his fitness prior to his reproductive success assuming he eventually generates many new poodles. Can you demonstrate correlation between reproductive success of his mother and father translates to his reproductive success?

    If he contains his mother and fathers genes and some random genetic variation what is the chance that this variation improves his fitness? What is the chance it degrades his fitness?

    If we cannot predict fitness then what are we really talking about?

    Another thing to consider besides adaptive traits, and which you might be more willing to accept is low fitness associated with heritable disease. A dog carrying alleles that predispose to debilitating disease (e.g. hip dysplasia) is expected to have fewer pups than a healthy dog. If the disease is heritable there is an increased risk of the pups getting the condition; a clearcut correlation.

    Predicting fitness is not that hard when you are willing to think about the underlying biology.

  12. walto: it depends what question you’re trying to answer.

    It depends what question WHO is trying to answer?

    I just gave you the question, if I am measuring the fitness of a shrew from the Jurassic, that is supposedly the original ancestor of all mammals on the plant, is the one shrew who became the first mammal the fittest any mammal can be, because it lead to every other mammal on the planet.

    That shrew has had 10 trillion descendants. That’s hard to beat. No matter how far anything ever evolves, it will never be more fit than that shrew. I wonder why we need evolution at all, if that shrew is the most fit any animal could ever be. Why didn’t evolution just stop then, because animals fitness has only been going downhill since. Whilst that lone shrew’s fitness keeps increasing.

  13. Corneel: Another thing to consider besides adaptive traits, and which you might be more willing to accept is low fitness associated with heritable disease. A dog carrying alleles that predispose to debilitating disease (e.g. hip dysplasia) is expected to have fewer pups than a healthy dog. If the disease is heritable there is an increased risk of the pups getting the condition; a clearcut correlation.

    Predicting fitness is not that hard when you are willing to think about the underlying biology.

    You have once again tried to take an extreme case of inability to breed, and drawn conclusions about fitness from that. Its like if you said, “Well, some men are born without a penis, don’t you think that affects their fitness? So you see, fitness is a useful concept. ”

    How does that make predicting fitness easy or real?

  14. phoodoo: You have once again tried to take an extreme case of inability to breed, and drawn conclusions about fitness from that. Its like if you said, “Well, some men are born without a penis, don’t you think that affects their fitness? So you see, fitness is a useful concept. ”

    How does that make predicting fitness easy or real?

    The accusation is that fitness is a tautological concept, that can only be established after reproductive success is known. Your example of men without penises (how do you come up with those examples?) demonstrates clearly that this is nonsense, and that there is real biology informing that prediction. You won’ t hear me saying that predicting fitness is easy, but it is definitely real.

    ETA: corrected mangled sentence

  15. Corneel: Predicting fitness is not that hard when you are willing to think about the underlying biology.

    You did say this right?

    Corneel: The accusation is that fitness is a tautological concept,

    You mean its not tautological to say that organisms that do not have the physical ability to breed, are less likely to breed?

  16. phoodoo: Corneel: Predicting fitness is not that hard when you are willing to think about the underlying biology.

    You did say this right?

    Yes, it was in response to Bill’s comment who seemed to be completely at a loss. I just wanted to point out that situations exist where predicting fitness is not hard (like your “men without penises”). I can see how you got the idea that I thought predicting fitness was always easy. To clarify my position: I believe that sometimes it is (e.g. developmental lethals), but usually it is not.

  17. phoodoo: I just gave you the question, if I am measuring the fitness of a shrew from the Jurassic, that is supposedly the original ancestor of all mammals on the plant, is the one shrew who became the first mammal the fittest any mammal can be, because it lead to every other mammal on the planet.

    No, fitness as used by evolutionary biologists is (in a simple case with diploid individuals and discrete generations) half the expected number of offspring. Taking probability of survival into account and expected number of offspring of those that survive. And it does not measure only those offspring that are the same genotype as the parent, and it does not measure subsequent generations.

    To figure out how genotype frequencies change in the population, we use the fitness but the equations are arranged so as to take the genetics into account.

    phoodoo will of course object, as usual, how do we know what the fitness is until after the fact. We might or might not — fitness is useful:
    1. To explore consequences of assumptions — if a genotype has a 1% higher fitness than another, and that continues to be true, how quickly does the gene frequency change, and where does it end up?
    2. As empirically measured on individuals of different phenotype values, to then predict what will happen when other individuals of the same population have their fitnesses determined by their phenotypes — the same “fitness surface”.

    It’s a useful concept. Yes, there are lots of possible violations of these simple models, and we try to deal with them one at a time, as science does in other realms.

    Declaring it circular, or useless will make no impression on biologists, who have seen for themselves that the concept is important.

  18. phoodoo: Corneel: The accusation is that fitness is a tautological concept,

    You mean its not tautological to say that organisms that do not have the physical ability to breed, are less likely to breed?

    But “physical ability to breed” is not one of the definitions of fitness is it? Expected reproductive output is, or let’s consider Allan’s proffered definition. Organisms that have a short time to maturation also have high fitness in Allan’s definition, but how is that tautological?

  19. Joe Felsenstein: We don’t know unless we know how many offspring it had (not how many ultimate descendants, how many immediate offspring).

    You just said its the expected number, not the actual number.

  20. phoodoo: I just gave you the question, if I am measuring the fitness of a shrew from the Jurassic, that is supposedly the original ancestor of all mammals on the plant, is the one shrew who became the first mammal the fittest any mammal can be, because it lead to every other mammal on the planet.

    Well, that question is easy to answer. No. Joe has indicated why.

    Maybe you’ll be happier with this, though. Using Allan’s definition DOES make a slogan like It is the fittest (entity, organism, allele, trait, species, etc) that will survive. IS tautologous. I think if one uses anything like the def he proposed for fitness, one will have to admit that what is expressed by the slogan is simply entailed by the definition. Hopefully that will be enough for you and we can all go home, because your shrew and penis questions are silly.

  21. Corneel: The accusation is that fitness is a tautological concept, that can only be established after reproductive success is known. Your example of men without penises (how do you come up with those examples?) demonstrates clearly that this is nonsense, and that there is real biology informing that prediction. You won’ t hear me saying that predicting fitness is easy, but it is definitely real.

    ETA: corrected mangled sentence

    Way too much mangling going on in this comment! X>{

  22. phoodoo: You just said its the expected number, not the actual number.

    For a genotype, it is the expected number. One way of estimating that is to average over all individuals of the genotype, in some large sample. That assumes that other loci, particularly nearby in the genome, have their genotypes uncorrelated with this locus. If we want to be more precise, at the cost of more work, breed an inbred strain where all individuals are the same genotype, then mutate one locus, and make another strain that differs just at that one locus. Then average the numbers of offspring of newborns for each.

    In my university department, there are a number of labs working on “experimental evolution” and their experimental design is very close to this.

    For an individual, which has already completed its life cycle, of course you can count the number of offspring (0 if it did not survive to reproduce).

    Lots of people are out there working on estimating how fitness depends on particular phenotypes (size, etc.). We don’t know in advance, though we may be able to model how that trait would affect fitness. After the observations, we know more. Not circular, can be used to predict how well other individuals will survive and reproduce, and what will happen in the longer run.

  23. walto: Well, that question is easy to answer. No. Joe has indicated why.

    Joe’s definition is the expected number. There was no expected number, the event already occurred.

    Tough to get a working definition, that’s for sure.

    And here I thought we already had one.

  24. phoodoo: Joe’s definition is the expected number. There was no expected number, the event already occurred.

    Tough to get a working definition, that’s for sure.

    And here I thought we already had one.

    See one comment earlier in the thread.

  25. Joe Felsenstein,

    Yes, I am seeing it, and we are back to square one with the ever malleable defintion. Its either expected or actual, and when it is expected and then becomes actual, it changes back to actual. Then of course we also aren’t taking into account the reproductive success of the next generation, because you said the only way to count the fitness of the shrew was to know how many offspring it had in its first generation, not its second or any after.

    So the shrew could have 500 that never gave birth again, and another could have only 1 and the 1 gave birth to 10,000 more, but the one who had 500 first but they all died shortly after birth becomes the more fit.

    Hey, just going by defintions…

  26. phoodoo: So the shrew could have 500 that never gave birth again, and another could have only 1 and the 1 gave birth to 10,000 more, but the one who had 500 first but they all died shortly after birth becomes the more fit.

    Hey, just going by defintions…

    Yes. All those things could happen, but the fitness of that individual would still be 500. If we want to know the fitness of a particular genotype we have to average over multiple individuals who have that genotype.

    Similarly, the “heads” probability of a particular biased coin is an expectation. You can estimate it by tossing it a lot of times, but of course that estimate is not perfect. Nevertheless, the concept is useful.

  27. phoodoo: Joe’s definition is the expected number.There was no expected number, the event already occurred.

    Tough to get a working definition, that’s for sure.

    And here I thought we already had one.

    Here, you’re simply confused about what an ‘expected value’ is in statistics. It doesn’t mean that somebody has to be around to ‘expect it’ first.

    https://en.m.wikipedia.org/wiki/Expected_value

  28. phoodoo: Behold the fittest organism that could EVER possibly be:

    LoL. But just think about all the offspring of the LUCA that did not leave offspring. It could in fact be the must unfit organism that ever lived.

  29. Mung: LoL. But just think about all the offspring of the LUCA that did not leave offspring. It could in fact be the must unfit organism that ever lived.

    Instead of giggling, you might explain some of these concepts to your buddy. Maybe he’d accept explanations from you. He’s apparently too stubborn to take them from anyone else.

    But maybe you’re too much of a team player to do this. Dunno.

  30. phoodoo: But sexual selection makes no sense if its just for the sake of preferred sexual selection.

    The whole point of sexual selection is that the reason they are being selected is for because the trait holds a better survival advantage.

    That isn’t “the whole point”, but it is an aspect of sexual selection.

    It is true that when it comes to explaining adaptations that enchance survival (as opposed to just reproduction), as in the ability to avoid predators, find food/prey, and survive the environment in general, in so far as sexual selection is invoked to explain such adaptations, there would have to be a correlation between the sexually selected trait, and survival ability.

    So if some female spider (say) is selecting among male spiders for their “dance moves”, if this sexual selection is to be invoked to explain general adaptations of the spider, like the ability to spin effective webs and hide from predators, then there would have to be some causal connection between these general spider abilities, and the actual dancing performance that the female finds impressive.

    As far as I am aware, there is some evidence for such correlations. But you are also right that there are examples where such connections are doubtful, and some where it could be argued they’re anti-correlated.

    If the only survival advantage is that its sexually selected, then even you hav to admit that is pretty circular.

    I would agree when and if sexually selected traits are simply assumed to also confer a general survival advantage.

    Generally I would say that being sexually selected isn’t a survival advantage at all, it is a reproductive advantage. Sexual selection might explain why a trait predominates in the population, but I don’t think it can be just assumed in any and all cases to also link to environmental adaptations that aid general survival. And if the response is that “it wouldn’t exist if it wasn’t positive for survival” that would be begging the question.

    I can easily imagine sexually selected traits that could be anti-correlated with survival, yet a specially forgiving environmental circumstance could allow the trait rise in frequency in the population, despite having a detrimental effect on the species as a whole.

    A case in point might be the many cases of flightless birds on isolated oceanic islands, where few or no natural predators can take advantage of the detrimental effect on survival that sexually selection for purely “aesthetic” traits might have.

  31. Rumraket: If the only survival advantage is that its sexually selected, then even you hav to admit that is pretty circular.

    I would agree when and if sexually selected traits are simply assumed to also confer a general survival advantage.

    No,no, quite the opposite. If a sexually selected trait is assumed to be selected because it is advantageous to have a sexually selected trait, THIS is what makes it circular. If it is assumed to also have some other usefulness, then its not circular.

    But what has been suggested, like for the peacock selecting mates because of their tail, the only explanation for the tail, is that that’s what females select.

    Its the :
    Q: Why do peacocks have large ornate tails?
    A: Because it attracts females.
    Q: Why does it attract females?
    A: Because females choose mates that will reproduce peacocks with ornate tails.

    I mean the whole story could easily just flip, like a polar magnetic shift. Females could suddenly just start choosing peacocks with ugly little tails. Then people would ask why they choose the ugly tails, and the answer would be, so they can have children with ugly tails that females will chose.

  32. Consider the difference between Absolute Fitness and Relative Fitness and the stories they tell.

    Absolute fitness reflects the numbers of offspring going on to reproduce.

    wA=vA fA or wA=(1/2) vA fA

    where

    wA is absolute fitness for genotype A.

    vA is the viability or probability that genotype A will become reproductively viable (i.e. alive and able to make offsring)

    fA is the fertility for genotype A.

    Well, the human population has experience a population explosion. vA has increased not because we genetically improved but because of the environment collectively of increased technology by humans. From an absolute fitness standpoint, one could say many genome types must be necessarily increasing in absolute fitness!

    Do medical researchers and geneticists monitoring the condition of the human genome think the human genome is improving. I haven’t heard as SINGLE study arguing we are improving, all the while absolute fitness of genotypes keeps going up, even the absolute fitness of sick genotypes associated with diabetes and other heritable diseases!

    Relative fitness? That’s more complicated, but it seems that if heritable diseases are becoming ever more persistent because of modern medicine keeping juvenilles with heritable diseases alive (like juvenile diabetes), then the sicker genotypes are persisting.

    A fundamental consequence of population genetic theory is that fitness is constantly improving. Does that look like what is really happening in the medical sense? Not to me, not to most people who actually are concerned with heritable diseases. That was one of the reasons John Sanford, after a successful career as a genetic engineer, went on to study the problem of the human condition from a genetic standpoint. As I studied John’s work and the work of people like H. Muller and Joe Felsenstein, it struck me there was a disconnect between the population genetic definition of fitness vs. the intuitive medical definition of fitness.

  33. stcordova: As I studied John’s work and the work of people like H. Muller and Joe Felsenstein, it struck me there was a disconnect between the population genetic definition of fitness vs. the intuitive medical definition of fitness.

    Two years later, a few thousands posts, and this just occurred to you?

    I think I have explained this disconnect at least 78 times.

  34. stcordova: As I studied John’s work and the work of people like H. Muller and Joe Felsenstein, it struck me there was a disconnect between the population genetic definition of fitness vs. the intuitive medical definition of fitness.

    Nah! Really? I always thought that words were used identically across different disciplines, conversations, situations, etc. Those evil evolitionists! How dare they not use a medical definition for fitness! Who do they think they are? Do they want to make sure that patients are healthy, or to try and model evolutionary events? We cannot have both in the whole planet. No differences across disciplines. That would be blasphemy.

  35. phoodoo: Q: Why do peacocks have large ornate tails?
    A: Because it attracts females.
    Q: Why does it attract females?
    A: Because females choose mates that will reproduce peacocks with ornate tails.

    I mean the whole story could easily just flip, like a polar magnetic shift. Females could suddenly just start choosing peacocks with ugly little tails.

    phoodoo, that is not a bad insight.

    In one hypothesis, called the handicap principle, the ornate tail is an honest signal of biological quality. Obviously, the ornament has to be costly, otherwise it can be copied by cheaters. So ugly little tails are no good.

    There are other hypotheses, but let´s first see how you like this one.

  36. Sometime in 2016, petrushka asked something about how fitness can be maintained even if the genome is being trashed and functionally compromised. I described above how absolute fitness can increase because viability increases due to improvement of human technology even though the genome is deteriorating from a functional standpoint.

    let me quote myself from 2016:
    http://theskepticalzone.com/wp/the-reasonableness-of-atheism-and-black-swans/comment-page-14/#comment-110270

    Here is a laboratory example of auto renormalization with the genome being utterly trashed but no reduction in “fitness” (reproductive capability, not really physical fitness in the medical sense).

    http://genetics.org/content/148/4/166

    In the 1950s, Wallace (1952, 1956) exposed caged Drosophila populations to continuous radiation accumulating to 250,000 r. The population accumulated a large number of recessive lethals, but its size was not reduced. This is perhaps not surprising in a species with a high reproductive potential. Nevertheless, although heavily mutagenized Drosophila populations showed no overt signs of genetic deterioration, they became weak competitors with nonmutagenized strains (Wijsman 1984). Thus, Wallace’s flies were indeed paying a price, but one that would have required a more rigorous environment to reveal.

    This was what I meant by autorenormalization. The genome can get trashed, but from a population genetics standpoint, it looks “fit” (note the scare quotes).

Leave a Reply