Thursday, 29 August 2013

Conversation with Oolon

I would like to thank Oolon for his kind words here.

There followed a bit of debate between the two of us. I am going to copy the back and forth here for future reference, so this isn't a blog post as such, just a reproduction of comments from another blog post for my own reference.

Green me to Oolon, orange Oolon to me.

The context of my first post is in response to Oolon finding fault with the Coyne/Pinker commentary on PZ's opinion of Evolutionary Psychology and what I saw as undue respect for Mark Hoofnagle's deconstruction of Ed Clint's analysis of Rebecca Watson's talk on Evolutionary Psychology.

I’m not sure PZ’s critique of Coyne amounts to much. There is some obviously fallacious stuff about PZ’s reply, for example when he says:

“When criticized, evolutionary psychologists love to run away from their discipline and hide in the safer confines of more solidly founded ideas. Here’s a perfect example:”

And then goes onto quote Coyne – who isn’t an evolutionary psychologist. So, perfect example how?
PZ still seems to think the premises of the field are fatally flawed – but he doesn’t display any knowledge of the premises. Maybe he’ll get round to that when he takes on Pinker, but I have serious doubts. His claim that EPs make observations from single populations isn’t particularly right. A study might use a particular set of participants – but any findings will only be advanced tentatively at such a stage (unless the authors are actually unethical). It’s only after some attempt has been made to show universality that the study’s notions are going to be taken seriously by others in the field.

PZ calls it a “cop out” of Coyne’s to not allow for the fact that PZ recognises that developmental plasticity and innate predilections do not necessarily rule one another out. But why should Coyne do that seeing as PZ has been saying stuff like “plasticity is *everything* and that should be the take home message for Evolutionary Psychologists” when talking about EP up to this point?

The rest – as Chas says in the comments – too vague.

Stephanie’s response isn’t much of a concern to defenders of EP unless she’s able to talk about the nature of the conclusions drawn from the papers that use students as participants. If they are wild – then yes, that could be a problem. If they are tentative or part of a wider body of knowledge then no problem. Pinker is, I guess, talking about the stuff that informs textbooks and high quality pop science books – replicated studies and meta-analyses. Stephanie is, I guess, talking about the bleeding edge of the field. She presumably misses the fact that those conclusions are likely to be tentatively held until some more effort to show universality is undergone.

(I would argue that reading actual scientific papers isn’t the best way to get your head round a field – you need to get the textbook knowledge and then read the papers).

Also, in failing to concede that PZ’s response in print was much the same message as the content of his section of the panel I think Stephanie does Coyne and Pinker a disservice.

Mark’s critique of Ed falls short of a debunk. Firstly they agree on a fair few issues, but most of Mark’s negative reaction to Ed seems based on his assumption that Rebecca accurately reports on her source material.

She does not. Most of the stuff she cites as examples is either somewhat or mostly at odds with what she says it is about. A couple of times she flashes up newspaper pages or scientific papers that have precisely nothing to do with what she attributes to them.

She also doesn’t seem to know which French king was the 14th Louis – but that’s a personal gripe.

Well it seems your criticism of the criticism all comes under the remit of “reasonable”. Whole point of this blog post is to pick out reasonable criticism as some people at a certain forum seem to like to make it their aim to paint all criticism as part of an evilz “call out culture” … This has the effect of allowing any criticism from PZ et al to be dismissed as part of this imagined pattern of behaviour and lowers the quality of debate. Just look at the reaction on Twitter to PZs criticism, doesn’t matter how measured he was (For him) the criticism will be dismissed. Even though Coyne’s was full of ad-hom and vague unskeptical calls to straw motives.

As for Rebecca and her presentation there were errors, she thanked Ed and modified her talk to incorporate the corrections and make it clearer she was aiming at Pop-EP not all of EP. So subsequent talks have these corrections in them. Not exactly the behaviour of a science denialist? for me that debunked Eds claim better than Mark Hoofnagle ever could have as science denialism is a pattern of behaviour not a one off talk with factual errors in it.

I can’t recall Rebecca thanking Ed, not to his face anyway. She claimed to have addressed a handful of the errors he attributed to her – the ones she couldn’t plausibly deny such as saying that a particular researcher was from a particular university when he wasn’t and so on. She certainly didn’t address all his objections. Her understanding of what amounts to pop and non-pop EP seems weird to me. Most of the time she just talks about EP whether her topic is pop or genuine psychology. The one time she mentions pop-EP is to indulge in the whole Pleistocene brain thing PZ and co mentioned so much.

The science denial charge? I don’t know if I would go that far but there are points in her talk where she seems to fundamentally misunderstand certain scientific principles. For example there are a couple of times where she basically runs through someone’s formation of a hypothesis, but does so mockingly and seems to joke with the audience about it. But why? Certain hypotheses might well be a bit wild. So what?

Also she seems not to know what an outlier is, she attributes stuff to scientists that actually come from journalists and – at times – thin air, she jokingly provides examples from myth to debunk claims that no one ever even made … and so on.

So I’m not sure I would be so bold as to call it science denialism per se – but it has a lot in common with someone like Lord Monckton opining on climate change – and that would be called an example of science denial by some of those who defend her talk. So colour me confused.

I don't know why I shy away from calling her a science denier myself - perhaps because I feel a bit differently about evolution and climate change than I do about social science. Put it this way - her misleading on the subject strikes me as every bit as severe as the likes of Ken Ham or Kent Hovind on evolution, or Monkton and the Heritage Foundation on climate.

And I think that should be clear, so whilst I do think that it's easier to criticise social science in many ways - just because of that I shouldn't shy away from calling her attempt at bad education what it is - which is effective denialism.

First comment here thanking “everyone” for corrections, even Ed. ->
For me it was an entertaining talk, she is not a scientist and was poking fun at pop-EP and the media’s representation and gullible swallowing of it not giving a thorough review of the state of research. That a few quotes were misattributed and some minor facts incorrect is not that important – was the overall thrust of the talk correct? Yes the media swallow pop-EP whole and more often than not its total bullshit. I’d imagine some of it goes back to sceptics like Ben Radford falling for the girls like pink crap. Rebecca wrote on that subject sometime back and Ben had to back down and apologise for falling for it.

So who is most to blame for EPs bad reputation in some circles? The media for presenting the badly written pop-EP stuff or Rebecca Watson for laughing at how awful the research they use for their stories actually is? I’d think if I was an EP researcher I’d want to get the media to report on it properly so RW etc have less ammo.

I'll note here that as far as I can see Rebecca does include "Clint" in her list of thankees...

Thanks Stephanie. I saw Clint’s post but as I’m traveling, I have no time to write anything up, so I’m very glad that you’ve done a great job of it. I’m actually giving this talk again tomorrow and I’m quite thankful to people who have given me notes and corrections. I even got a few good ones from Clint! He’s absolutely right that I misspoke in regards to Kruger’s affiliation (it’s U of Michigan, not Chicago, that should be embarrassed) and in regards to the favorite color study being given to Chinese people in the UK, not in China. Also, the “Why People Have Sex” study was not all white middle class women – it was only about ~60% white (and ~20% Asian.) I think I’ll note instead that the study involved 96% 18-22 year olds, all of whom were psychology students at University of Texas Austin, and among the women 27% of whom had never had sexual intercourse. More accurate and also more ridiculous.

There are other bits and pieces Clint got wrong but at a glance I think you’ve covered the bulk of the problems here.

... though I personally find her general tone so utterly begrudging it's hard for me to credit it. I don't think offering qualified and begrudging thanks on a third party's blog counts for much, particularly when it comes with a dismissal of the vast majority of the things he was trying to bring her attention to. However, it technically serves for the purpose of argument I suppose.

Her talk was almost all misconstrued. I had a review of the first 5 mins here:

Then up to 12 mins or so here:

And then later to 19 mins here:

I didn’t get to the “pink is for girls” stuff. But in short:

Dr Yazhu Ling and Prof Anya Hurlbert are not evolutionary psychologists. Prof Hurlbert is a neuroscientist, Dr Ling is eclectic. Their study has some interesting factors to consider but it has not been warmly received by EPs. Their conclusion is that the phenomena is definitely sociocultural though they speculate as to a possible innate factor. It’s also part of a wider body of work as the two have collaborated on other papers to do with sight and colour perception. It isn’t pop-EP – it’s cog-neuroscience with some speculative analysis that might be of interest to EPers – but doesn’t seem to have got much attention (see Ed Clint’s dismissal of it, for example).

Personally I think he’s too harsh – study may have something to say about effects of colour blindness in that for 10% of men red and green are somewhat ambiguous.

Rebecca chucks it for wrong reasons. Sociocultural shifts in regard to fashion are irrelevant to notions of innate tendency, marketers may even be honing in on actual phenomena. Or they may not.
So, seems like a fuss about nothing much to me.

You say they are not “evolutionary psychologists” … One of the first Google results for their names gives
Evolution may have driven women’s preference for pink, according to the study published today…. [big jump] …. However, Professor Hurlbert says she could only speculate about the universal preference for blue: ‘Here again, I would favour evolutionary arguments. Going back to our ‘savannah’ days, we would have a natural preference for a clear blue sky, because it signalled good weather. Clear blue also signals a good water source’, she says.

Hasn’t got much attention? Maybe by the true EP researchers but that was not the focus of RWs talk – pop-EP was. Widely reported as “evolutionary psychology” even by the benighted Ben Goldacre ->

Maybe the answer to why EP has such a bad rep is that even sceptical top cats such as Ben misreport research as being by EP researchers. (Can’t remember him being called a science denialist) Either way RW was critiquing the media’s reporting of EP, pop-EP, and this was most definitely reported as EP.

Ben Goldacre benighted? Sainted more like. Then again, I read the Guardian and they're unlikely to slag their own correspondent.

It’s not one of Ben’s better columns. He doesn’t point out that it is they, not he, who produce a sociocultural explanation and there’s a difference between an evolutionary psychologist and a psychologist who mentions evolution. Even if you take that as irrelevant why would a paper providing both sociocultural and evolutionary explanations be deemed EP – especially when those evolutionary explanations are advanced tentatively? It is – at most – an eclectic paper.
Nor does he mention that – in social science at least – it’s expected that researchers indulge in a speculative analysis having drawn a conclusion.

This is why the press release you mention uses words such as “may have” and “could only speculate”.

To see what impact it has had on the field you’d need to see how it has been cited. It hasn’t been cited much, and those papers that do cite do not seem to have much to do with EP (though frustratingly I can’t the page to work properly, so maybe later papers tell a different story):

I'm not entirely happy with that answer, I'm a bit frustrated at the way Anya Hurlbert and Yazhu Ling get trotted out every time someone wants to talk about a "clearly ridiculous EP paper".

I do think there are some poor things about the paper - it's not clear to me what the hypothesis is, I am assured that the effect size is too small to say much about, the trichromacy they discuss, whilst interesting, does not lead me to believe that women really benefit from distinguishing between red and green to an effective extent (colour-blindness aside), nor do I see how distinguishing between shades of red could help in gathering leaves (though having done a bit of foraging myself I will admit that you soon become quite aware of very subtle variations in green that may result from red pigments, like distinguishing between various species of Orache and so on, so perhaps this isn't as stupid as it seems).

Anyway, these aren't the main objections raised by people who want to challenge EP (and I'm not saying Oolon is such a person or that he makes all of these judgements). They tend to say:

A) It doesn't leave room for an alternative sociocultural explanation.

It does, explicitly, in the final analysis. In fact they anticipated a sociocultural shift according to both gender and whether or not the participants were Han Chinese or not. In fact the sociocultural stuff is expressed as a (pretty much) certain phenomena, whereas the evolutionary explanation is prefaced with qualifiers such as "we speculate" and "might be".

B) They have this notion of "pink for girls" which is a concept that we have seen change in recorded history.

To be fair, they admit that the notion is why they came up with their experiment, but that is just admitting to your influences. The experiment tests something different, which is what colour people are drawn to and what colour they deem "fitting for a particular gender". That the notion that a particular colour befits a particular gender inspired an investigation into what colour people are drawn to with gender as the independent variable means very little.

"My favourite colour is pink."

Is compatible with either:

"I would never dress a little girl in pink." /or/ "I always dress my little girl in pink."

Seems clear to me.

C) This sort of study is typical of EP.

Not really. As I said earlier neither Prof Hurlbert or Dr Ling identify as evolutionary psychologists. If their body of work is taken as a whole it seems more that they are investigating lots of phenomena to do with vision and so this is just one aspect of a bigger picture.

As I said to Oolon, that study hasn't been warmly welcomed by evolutionary psychologists as far as I'm aware. I haven't seen it positively cited much, let alone in any sort of decent introductory work.

So it's not typical EP - it's barely representative.

D) It used to be "pink is for boys" - "pink is for girls" is just marketing.

I know we should all hate marketers because they keep on persuading (or trying to persuade) us to engage in empty consumerism. However, it may be that, on occasion, they employ psychology effectively in order to make a better sale.

As such the notion that something is not a genuine phenomena due to marketers being involved is fallacious. Marketers are probably a little more likely to understand a psychological phenomena than laypeople, even if they put it to the service of a Dark Art.

My feelings at this point - personally I'd like to see some attempts at replication of the study. I think it's reckless to say its up to much in and of itself, but it's equally reckless to say that the phenomena it seeks to explore does not exist. I don't think it's a bad study, but I don't think it's great either and the paper is poorly laid out as far as I can see. Either way - it's not what I think of as representative of EP.

I agree its not an EP paper from what I can see. Point is its presented as such in the popular press, also the speculative evolutionary part of the psychology paper is presented as fact by the papers. Hence pop-EP has a bad rep and hence why RW attacked it. A lot of the criticism seems to be talking past each other as there is clearly pop-EP that is very worthy of attack. To then take that justified attack and say its an assault on all EP is a stretch. Especially when PZ said in the CONvergence panel that he sees EP as a dead end and it should be scrapped! Seems some ppl are happy to say the whole field is a dead loss… You are not short of people actually saying its all bunk :-)

So what is good EP? I know PZ says he’s never seen a good paper, but what is EPs best foot? Where has it clearly demonstrated an innate behaviour that is consistent across multiple populations? Where is the null hypothesis of culturally conditioned behaviour ruled out? What practical benefit has been achieved through EP research? I’ve seen nothing yet myself so those are all questions to me that undermine EP – although again I’ve not done the reading to be able to assert with any authority. For other fields its not even in question though, so do I take the relative infancy of the field into account or assume that PZ is right and its all hogwash?

Followed later by:

Hehe I certainly see why ppl focus on pop-EP … Its a pretty easy target. I, err, came across this today. Discussed as ridiculous by a number of feminists on Twitter etc.

One point stuck out for me I was pretty sure the suckup theory was bollocks and proven so ages ago. Although seems while most studies have shown no correlation one did show greater jiz retention. Obviously that is not necessarily going to lead to greater fertility, but it might. Then a recent one seems to totally blow it apart -

Not to mention the WEIRD and methodological issues in the paper. Seems very easy to find crappy EP day to day on the internet. Not good EP.

Apologies for delay in reply –

(I pimp this blog for a bit, which needs no repeating here)

“So what is good EP?”

Stephen Pinker’s How the Mind Works, The Language Instinct.
David Buss’ Evolution of Desire, EP the new Science of the Mind.
Sarah Blaffer Hrdy Mothers and Others.

“I know PZ says he’s never seen a good paper, but what is EPs best foot?”

How would he recognise one seeing as his understanding of the field is limited to broad misconstruction?

But as I say in the blog, it isn’t particularly useful to try and understand a field from the papers alone. Until there is some realisation on the part of the reader as to what the underlying ontology and epistemology are.

For example …

“Where is the null hypothesis of culturally conditioned behaviour ruled out?”

Why would that be a null hypothesis? Sometimes psychological phenomena are a combination of sociocultural and innate inputs. Sometimes neither (as is the case with individual difference).

When you ask questions like that it’s clear you don’t know much about psychology.

An easy example – on average Inuit wear heavier clothing than equatorial tribespeople.
This is sociocultural, but also a matter for EP, because we have natural inclinations to keeping our bodies at a comfortable temperature.

“Where has it clearly demonstrated an innate behaviour that is consistent across multiple populations?”

Cross culturally children imitate the facial expressions of others within minutes of birth.

Cross culturally people crave energy rich foods high in sugar, protein and/or fat, and are adverse to foods that are bitter.

Cross culturally women bear the costs of childbirth and are more discriminating in regards to sexual habits than men are.

Whilst preferences for weight in a female mate vary cross-culturally preferences for a certain waist-to-hip ratio are shown.

Cross culturally and on average men are shown to outperform women in tasks relating to thinking about 3d space – whilst women outperform men in tasks involving linguistic cleverness.

I mean, how long and varied do you want this list?

“What practical benefit has been achieved through EP research?”

What about bolstering the following line of argument:

People are mammals and we should think through the ethical implications of the fact that it is women who bear, nurse and disproportionally raise children. One ought not to assume that the default human being is a man and that children are an indulgence or an accident that strikes a deviant subset. Sex differences can be used to justify, rather than endanger, woman-friendly policies such as parental leave, subsidized childcare, flexible hours and stoppages of the tenure clock…
- Steven Pinker, The Blank Slate, page 358.

” For other fields its not even in question though, so do I take the relative infancy of the field into account or assume that PZ is right and its all hogwash?”

I would say neither.

An argument in social science is what it is. I don’t particularly invest in stuff like whether or not a female orgasm leads to increased chance of pregnancy. That’s one hypothesis amongst many. If it fails I don’t see that as a problem for EP – it’s just a hypothesis.

If people like PZ want to throw out the baby with the bathwater because some stuff strikes him as bad or unwholesome (and note that he has yet to provide much example of something that can be said to be both representative of the field and irrefutably nonsensical) then presumably he’ll be able to offer a sociocultural explanation for things such as the phenomena I cited earlier as cross-cultural.

That's the conversation as it stands. I wish I had mentioned a couple more things.

Yes, plenty of people sneer at EP. Is this phenomena different to the plenty of people who sneer at quantum physics, or evolution, or climate science, or cosmology? If so how so?

I would also love to know why Oolon is so much more impressed with the participant samples and methodology of the article on the science direct site than the one on the evolutionary psychology journal site - because I bet it's as easy to critique despite his claims that it blows the EP study apart. I doubt it does - to my eye they seem to be taking the problem with roughly equal seriousness and come to a few similar conclusions. More research needed.

1 comment:

  1. Interesting exchange. A couple quick points.

    I did not formally reply to Mark Hoofnagle because his criticism was of such low quality. That was very disappointing.

    re: color preference and gender. There is evolutionary social science work here, and I saw a few presentations along these lines at HBES. But none of them (or other EP analyses I have seen) ever say "pink is for girls, blue for boys". Instead, it seems women are more sensitive to the red-green axis and men to the yellow-blue; in some cases, this does seem to link to preferences. Many of these studies are not done by people with anything to do with EP, but rather psychologists/neurologists who study vision, but the data remains useful to evolutionary consideration.
    This could help explain partly why a marketing effort might succeed, even if much of the specificity is arbitrary (nobody has ever shown human females in general prefer pink, or males blue).

    Lastly, I think the context of what Watson is on record as saying makes the claim of "mere pop psych" untenable, and a little insulting. It is quite clear she uses the terms interchangeably, even in the same conversation. It's also wrong for another reason: it's just plain false. The depiction of EP in the media isn't wall-to-wall glorification of Kanazawas and Hurlburt/Ling's: it is in the majority negative criticisms about EP being wrong, silly, or pseudo.

    Perversely, folks seem to read the such negative accounts (which may not even refer to any evolutionary psychologist) and accuse EP of presenting itself badly. Astonishing.