More Bunk Science

Perhaps since the paper's authors wrote ... They were aware that symmetry is not the be all and end all of what is considered attractive. Also that was not the point of the study at all, which was concerned with co-operation in a prisoner's dilemma game.Has to be said BD, in this case the one who seems to have come in with a preconceived idea and tried to shoehorn the facts to fit it seems to be you. Also as others have pointed out, criticising a scientific study based on how it is reported in the newspapers is almost always an exercise in futility as the media has given you a pre-built strawman to beat up on.TTFN,WK

Maybe you should read the paper, from what you write here it is obvious you haven't so far.TTFN,WKEdited by Wounded King, : No reason given.

I'll help you out, there is a pre-publication draft of the paper available on the Edinburgh University website here. Except in this case they had several very specific hypothesis driven traits, circulating testosterone levels, digit 2 to digit 4 ratio, facial masculinity, fluctuating asymmetry which there was already considerable evidence suggesting they were related to cooperative or aggressive behaviour. They also looked at other variables including age, which city in the study the subjects came from and the behaviour the subjects expected from their opponent prior to them actually taking part in the game.The scenario you describe is the sort of approach which might turn up an association but is very unlikely to throw up one that isn't a false positive. That is why it is beneficial to go into a study with a defined hypothesis rather than the sort of fishing expedition you describe.As has been suggested you should try and get a basic understanding of statistics and then you might understand why not looking at hundreds of traits actually makes any association found more robust.And also you really should read the paper before making any more comments about it.TTFN,WKEdited by Wounded King, : No reason given.

Care to make some specific coherent criticism rather than this vague hand waving? Why do you not consider the significance of the results in their study to be robust? I don't see the distinction you are trying to draw here, you seem to be saying that even allowing that this correlation does exist it doesn't mean that more symmetric people are less co-operative, just that less symmetric people are more co-operative because they are more stupid? Can you point out to me how those 2 things are actually different?The only difference I can see is that you have added an entire additional layer by tying things into intelligence, which goes considerably further beyond what the data will support than anything in the paper.As to it being an evolutionary artefact, can you tell me where in the paper they make any such claim? I don't find any mention of evolution in the paper outside of the references. In the press pieces certainly, but not in the paper itself (at least the draft one I have access to at home).Also on a point of information, they aren't addressing the 'Nobel Prize committee' it is rather a meeting that will be attended by several Nobel laureates in economics.TTFN,WK

Presumably like getting creationists to tell the truth and not just make up stuff about evolution?TTFN,WK

But you predicated it on what can only be called a lie. When you are arguing from a completely false basis, as you have been for pretty much this entire thread, how can we be expected to take anything you say seriously? Can you show where the paper makes such a claim? As it happens, I can show you studies that say that (see below) but it isn't germane to what the paper you have yet to coherently criticise says. That isn't a reason, it is a question and one that you certainly haven't tried to find out the answer to, since there is significant evidence suggesting that the answer is yes, which rather undercuts your argument. Rural Belize is one place (Waynforth, 1998) and you apparently get more sex at college if you are more symmetrical (Thornhill & Gangestad, 1994).Maybe you should do some research for yourself before you start telling us all how it is and what evolution says. You would be less likely to be completely wrong all the time and look less like you just make stuff up and lie.TTFN,WK

Does this mean that we might eventually see you actually return to the original paper and the statistical analysis you were going to give us as to why its results were weak?As far as I can tell you are the one who has done their best to go off topic repeatedly, right from the OP when you managed to completely misrepresent almost every aspect of the paper that you claimed to be criticising. Seriously? I can hardly keep track of the number of times in this thread that you have stated things that the paper claimed according to you which were not in the paper. But fine, you were misrepresenting 1.61803's position rather than the original paper's, it gets hard to keep track sometimes.So basically you agree that you have stopped discussing the paper? As I said, which of us is it that is taking things off topic? Could you rephrase that so that it isn't a complete misrepresentation of both what the papers I presented say and what current evolutionary-developmental theory suggests? Once again you are constructing a strawman from your own fevered imaginings. You mean you'll bitch about it, misrepresent it and make lots of completely unsupported assertions about it? I find that all too easy to believe but I'm not sure why you think we would want you to do that with yet another paper when we are having such a hard time getting you to actually meaningfully discuss the first one. All this suggests to me is that once again you aren't bothering to actually read the papers, or if you are you aren't understanding them. The commonly understood reason was in the introductions to both of those papers, that fluctuating asymmetry is a reflection of the developmental stability of an individuals genotype.Therefore FA is a proxy measure for more developmentally robust, and typically fitter, genotypes. If low FA is indeed an honest signal of higher fitness then it is reasonable that prospective mates will favour individuals with lower FA. Both introductions are also full of references to the empirical studies that support this hypothesis. At last something we can agree on.TTFN,WKEdited by Wounded King, : No reason given.

This is demonstrably the case in almost any population where such questions are relevant, i.e. sexual populations with non-random mate choice based on visual cues. Does every female only mate with the most symmetrical male in the planet? Obviously not. There is a continuous spectrum of fluctuating asymmetry and almost certainly a similarly discontinuous spectrum of traits allowing both the discernment of such and a preference for mating with such individuals. That makes no sense, it is in fact the complete opposite of what we have been discussing. Well almost, assuming that there is a spectrum of preferences in the population those who have a preference for low FA will be mating preferentially with higher fitness individuals than those with a preference for high FA, assuming that low FA is an honest signal of developmental stability. That means that natural selection will be tending to favour a preference for low FA. That isn't an acquired characteristic in the sense it is usually understood though. Quite possibly not, it is probably a complex polygenic trait, there might well be an element of cultural inheritance in mate choice as well. What we may well find is that such preferences are heritable to a degree however. There is considerable evidence for the heritability of some mating preferences in other species so there is no reason why choosing low FA shouldn't be one. No they don't, it is as simple as that. You have just chosen to ignore all the years of background work that provide most of these logical connections, which is why this thread has degenerated into a series of rabbit holes with you questioning every paper I cite and asking how each one knew about some aspect of the hypothesis, which leads us to a further back paper where we start all over again.Why not actually go back to the original paper and make a coherent crtique of it, such as the statistical reasoning which you claimed was so flawed but didn't see fit to actually explain how it was flawed.TTFN,WK

Because previously you haven't coherently articulated them. Previously you simply stated that their results made it 'statistically impossible to draw conclusions' which is an incredibly strong statement and one that should really be supported by more than some overlapping standard deviations. No it doesn't, although it does make them overlapping in terms of standard deviation. If you graph these out then the standard deviation's do overlap considerably, but the more important measure would be the standard error. Standard error is calculated as ... ... where SE is the standard error, sd the standard deviation and n the number of subjects in the study. So the SE for the FA would be ...Defect Cooperate Which I'm sure you appreciate is considerably smaller and which produces no overlap.If you think there is a flaw with their calculations giving them a p-value of 0.015 for the differences between the 'defecting' and 'cooperating' populations then what do you think it is? The fact that you don't think such overlapping populations should be significantly different statistically doesn't really count for much. A p-value of 0.05 is a commonly used cut off for statistical significance, it means that there would only be a 1 in 20 chance that the results would be as they were from the study despite there not being any real difference in FA between the populations. Given the actual p-value the chances for the results being due to chance in this case are much lower, around 1 in 66.Ideally we would have access to the raw data for the study so we could perform our own analyses and confirm the statistical significance of the relationship. In the absence of the raw data I'm not sure what your basis is for claiming that the statistical significance they calculate doesn't exist? Do you think they are trying to be fraudulent or that they simply used an unsuitable statistical test? The t-test seems to be appropriate to me for comparisons between two populations as in the study. I don't see how this is true, the study is designed to test the hypothesis, in fact it looks at several distinct hypotheses. That is what the results showed, if you have an actual critique of their statistics that casts doubt on their significance values then present it, otherwise you are just giving us your opinion and expecting us to give it the same weight as statistical analysis. The entire point of statistical analysis is to test this contention. I would also question whether this is 'a small enough study' there are frequently studies with truly dubious levels of n, i.e. less than 10 or 20. One thing to bear in mind is that the necessary sample size is really dependent on the strength of the effect being studied. If the effect is strong enough it will be detectable even in a smaller study.As to the last bit, that seems more like what you were advocating before, looking at hundreds of different variables. Going back to our p-values, it is easy to see that we might expect to see as many as 4 independent variables form a set of 100 giving us results which appear statistically significant at the p<0.05 level even if there was really no difference between the 2 populations. That is why looking at specific hypothesis driven variables is preferable to the sort of fishing expedition you were advocating. Well that is another issue and one that has long been thought to plague evolutionary psychology. Sadly you can't due to what is known as publication bias the data exists in tucked away lab books and long forgotten folders on hard drives in labs all over the planet. It absolutely does. There is a lot of research, especially negative research, that never gets published. Yes, people who know anything about science publishing.TTFN,WKEdited by Wounded King, : No reason given.

Well duh! You have said so several times, you just don't seem to have any actual rationale for saying it, beyond your own prejudices. Clearly your ability to understand statistics is non-existent. The whole point is that the p-value shows us that despite the overlap in the raw measurements, in terms of standard deviation, the populations are still significantly different in their means at the P<0.05 level. What margins of error are you talking about if not the standard deviations from table 1? Why do you choose to ignore the fact that the standard error is considerably smaller than that. Because this fact is incidental to whether or not this paper is bunk science as you claimed. If they have showed a significant association but you disagree with the interpretation then fine. You want to put it down to intelligence and claim it doesn't really reflect cooperativeness, I'm not much bothered. These elements are interpretation and discussion of the results, they aren't the results themselves. There is even research out there that would support your view since FA has also been shown to negatively correlate with some measures of intelligence (Burlow et al 1997). I haven't really found any equivalent studies looking at whether IQ correlates with defect/cooperate behaviour in the prisoner's dilemma. The only one I have come across looked at groups playing the game over a repeated series of games rather than individuals in a one off situation, in that case they found that groups of higher intelligence seemed to cooperate more (Jones, 2008). Except that it wasn't, it was created more than 50 years ago by game theorists and is one of the most studied game theory scenarios. It may shock you to learn this but scientists don't get to dictate when and where their research gets published. They have to get it through peer review and past journal editors. There are a variety of factors which have lead to what most people would agree is an invidious situation, a large part of which is that scientific publishing is by and large a commercial enterprise. And I want you to stop making things up, but I guess you can't always get what you want.TTFN,WKEdited by Wounded King, : No reason given.

No it isn't. Yes there is, because that is what the data strongly suggests. The chances are pretty darn high that you would see significant differences in the means in the same way from a larger population. There is a small chance that the trend might disappear or reverse, but the whole point of calculating the statistics and specifically the p-value, is to give us a measure of our confidence that it won't. You don't seem to understand what statistics is for.TTFN,WKEdited by Wounded King, : No reason given.