Are most experimental subjects in behavioral science WEIRD?

Note: here is a follow up post.

My supervisor David Spurrett and I have a commentary on an important paper - "The weirdest people in the world?" (pdf) - in the most recent edition of Behavioral & Brain Sciences. The authors of the paper, Canadian psychologists Joseph Henrich, Steven Heine and Ara Norenzayan, argue that most experimental subjects in the behavioral sciences are WEIRD - Western, Educated, Industrialized, Rich, and Democratic - and thus weird - not representative of most human beings. And this, if true, is a very serious problem indeed. Behavioral scientists (anthropologists, psychologists, behavioral economists and so on) are often interested in explaining the brains, minds and behavior of Homo sapiens as a species. (Some scientists, of course, are only interested in understanding specific cultures or what makes us different, but one important goal of the behavioral sciences has long been to explain universal human behavior). As evolutionary psychologists John Tooby and Leda Cosmides have put it, they "seek to characterize the universal, species-typical architecture of [the information-processing mechanisms that generate behavior]".

But... Henrich and his colleagues review a large body of literature that seems to show that, across several domains, Western undergraduates - the workhorses of the behavioral sciences - are extreme outliers. In other words, if they are correct, most of the data behavioral scientists have used to test hypothesis and to drive theorizing derives from subjects who are possibly the least suited for generalizing about the human race. Take as an example the Müller-Lyer illusion. In the diagram below, the lines labeled "a" and "b" are exactly equal in length, but many subjects perceive "b" as longer than "a".

This finding (which goes back all the way to 1889) has been used to make deductions about how the human visual system works. The Wikipedia article on the illusion, for example, states that one possible explanation for the effect is that "the visual system processes that judge depth and distance assume in general that the 'angles in' configuration corresponds to an object which is closer, and the 'angles out' configuration corresponds to an object which is far away". Plausible enough. Except that for some people - San foragers, for example - the illusion does not exist, and in many other non-WEIRD societies the effect size is significantly smaller. Henrich and his colleagues cite the work of Segall et. al. (1966), who worked out the magnitude of the illusion across 16 societies by varying the relative lengths of "a" and "b" and then asking subjects to indicate when they thought the lines were equal. The percentage by which "a" must be longer than "b" before the lines are adjudged equal - what they call the "point of subjective equality" (PSE) - varies substantially between subjects from different cultures - and, importantly, WEIRD-subjects are extreme outliers. The results are summarized in the following graph:

Both WEIRD adults and children (aged 5-11) require "a" to be 18%+ longer than "b" before they're perceived as equal, but for the San and South African miners, the illusion simply does not exist - their PSEs are not statistically distinguishable from 0. Why this difference arises is unknown, but Segall et. al. claim it is due to WEIRD people's visual systems developing differently because modern environments expose them to ("unnatural") shapes like 'carpeted corners', thus calibrating their visual systems in a way that favors the emergence of the illusion. Whatever the true explanation, however, it is clear that it is not permissible to use the existence of the illusion among WEIRD subjects to make inferences about the visual system. This is especially true since the San subjects were hunter-gatherers, just like all people for the vast majority of human evolutionary history. Given that species-typical features of the visual system would have evolved in this period, it is particularly telling that PSE seems to be positively correlated with the 'modernity' of the societies in question. (Warning: this is an "eyeball" observation; I haven't done a proper statistical analysis. Caveat emptor).

This is one example from an extremely long paper, but it conveys a flavor of the kind of evidence the authors present. (For much more, see "We agree it's WEIRD, but is it WEIRD enough?" over at Neuroanthropology). Having read the article very carefully, and despite some concerns, I think Henrich, Heine and Norenzayan are right: the Western undergraduate is often unrepresentative of humanity, and the behavioral science literature needs a lot of fixing as a result. (Most obviously, we need far more large, highly-powered, globally representative, prospectively designed, cross-cultural studies). Serious as this is, unfortunately, it gets worse... Since David and I worked extremely hard to present our argument clearly and concisely in our commentary (pdf - our piece starts on p. 44 of the pdf, paginated by BBS as p. 104), and I doubt I could improve on it, what follows is a slightly edited - simplified and somewhat de-academicized - version of the meat of our argument. (Note: each issue of BBS consists of a "target article" - in this case, Henrich et. al. - and 20 or so short peer-commentaries).

Henrich et al. underplay – to the point of missing – that how the behavioural sciences research community itself is constituted introduces biases. That the subject-pool of behavioural science is so shallow is indeed a serious problem, but so is the fact that the majority of behavioural researchers are themselves deeply WEIRD. People in Western countries have, on average, a remarkably homogeneous set of values compared to the full range of worldwide variability (Inglehart & Welzel 2005), and the data Henrich and his colleagues present suggest similarly population-level homogeneity in cognitive styles. Moreover, academics are more uniform than the populations from which they are drawn, so it is likely behavioral scientists are even WEIRDer than their most common subjects. Henrich and his colleagues review a bunch of studies and experiments that did not strike those who designed and conducted them as focused on outliers. Intelligent scientists acting in good faith conducted, peer-reviewed, and published this research, in many cases honestly believing that it threw light on human nature. This forcefully illustrates the power of the biases on the part of researchers themselves. It also suggests that, besides widening the pool of subjects, there are significant gains to be made by broadening the range of inputs to the scientific process, including in the conception, design, and evaluation of empirical and theoretical work. Given that diverse groups are demonstrably better at some kinds of problem solving, as things stand, the WEIRD-dominated literature is robbed of potentially worthwhile perspectives, critiques, and hypotheses that a truly global research community could provide. Clearly, simply increasing the number of behavioural sciences researchers will, in general, be beneficial. Our key contention, though, is that the marginal benefits of additional Western researchers are much smaller than the marginal benefits of more non-Western researchers, among other things, just because they are non-Western.

The non-Western world, in short, can contribute not only additional subjects to experiment upon – the main focus of the target article’s recommendations – but also additional researchers, with novel perspectives and ideas and who are less affected by WEIRD biases. (Naturally, these researchers will have biases of their own. Our claim is not that there is someone who consistently knows better; it is that diverse groups of investigators can avoid some kinds of error.) Clearly, these researchers will have to be educated, will likely be middle class, and, since science flourishes in politically open societies, they will tend be concentrated in liberal countries. Nevertheless, additional non-Western researchers, even if they are educated and relatively wealthy, could be a boon to the behavioural sciences.

A direct and powerful way to remedy both sources of bias – too many WEIRD subjects and too few non-WEIRD researchers – is to foster research capacity in the non-Western world. Non-WEIRD researchers tend to study non-WEIRD subjects, so increasing their number will deepen the subject pool and widen the range of inputs to the scientific process at the same time. Building research capacity, however, should not merely involve collaborations led by WEIRD researchers; it should aim to generate studies led and initiated by non-Western researchers. Committed and long-term inter-institutional collaboration between Western and non-Western universities focused on remedying the deficits in the behavioral sciences literature should include internships at Western universities for non-Western researchers, stints at non-Western universities for WEIRD researchers, and extensive student exchange programs (especially for graduate students). Unlike many existing scholarship and exchange programs in the sciences, a key point of the necessary programs should be for the learning to proceed in both directions.

----------------------------------
Henrich, J., Heine, S., & Norenzayan, A. (2010). The weirdest people in the world? Behavioral and Brain Sciences, 33 (2-3), 61-83 DOI: 10.1017/S0140525X0999152X

Meadon, M., & Spurrett, D. (2010). It's not just the subjects – there are too many WEIRD researchers Behavioral and Brain Sciences, 33 (2-3), 104-105 DOI: 10.1017/S0140525X10000208

Depictions of violence in rock art

A street fight, via Wikipedia.

To understand the phenomena of murder, war, genocide, and other forms of human intraspecific violence, we need to know whether to invoke evolutionary biological explanations or restrict ourselves primarily to socio-cultural theories. If the incidence of violent conflict was high and recurrent for a substantial period during human evolution, and given that being killed drastically reduced fitness and killing may have increased it, then strong selective pressures would have favored physical and psychological adaptations to violence. Conversely, if interpersonal violence was rare or nonexistent until much more recently – until the rise of agriculture about 10,000 years ago, say – not enough time would have elapsed for natural selection to have forged significant new adaptations, and socio-cultural explanations of violence would thus predominate. (It should be noted, though, that recent human evolution has been very rapid, so this judgement may have to be revised in future as more evidence comes in). More precisely, whether adaptations to violence exist or not depends on the intensity of the selection pressures and their duration, and the intensity of the selection pressures is in turn a function of the frequency of violence and the magnitude of its impact on fitness. Thus, to determine the plausibility of positing traits that are adaptations to violence we need to know: (1) how frequent violence was, (2) whether it was recurrent in human evolutionary history and (3) how large its impact on inclusive fitness was.

Determining (3), of course, depends in part on the values we assign to (1) and (2). Being killed before reproduction obviously reduced fitness to zero, and being killed after reproduction eliminated all the kin altruism the individual would otherwise have engaged in. The impact on fitness of being injured depended on the severity of the injury, but it seems clear it would have been negative and serious. What we need to deduce the magnitude of (3) over human evolutionary history, then, is sound empirical estimates of (1) and (2). Unfortunately, however, these estimates are extremely difficult to make because the available evidence is sparse and often ambiguous. Broadly speaking, there are two lines of evidence available to us: studies of contemporary hunter-gatherers and the paleoanthropological and archeological record. There are several controversies around both lines of evidence, but for this post I'll focus on one type of evidence from the archeological record: depictions of violence in rock art.

A beautiful example of such a depiction is a San pictograph "Veg 'n Vlug" (Afrikaans: "Fight or Flight") that is near Clanwilliam in the Cederberg of South Africa. (Note: I've used the Auto-Level feature in Paint.NET to bring out some of the details):

John Parkington describes (large pdf; pp. 62 - 65) the scene thus:

The fight element is created by painting around a small recess in the rock surface to give the impression of a small cave from which a group of humans peer, one of them shooting arrows. A second group of humans, arranged as a procession and depicted apparently moving along a pair of red lines, face the cave occupants and also shoot arrows. From the ‘cave’ several people, most of them male, flee along more pairs of red parallel lines. One human figure, clearly lying prone is connected by these same lines to a strange seated  figure holding the end of the lines, neither of them directly connected with the cave itself. From the neck of the strange seated figure a single red line leads to another small figure with upraised arms.

Contrary to the hypothesis - favored by neo-Rousseauians like Brian Ferguson - that human evolutionary history was entirely (or largely) peaceful, then, we have at least an existence proof of such violence. Or do we? Ferguson has argued that pictographs seemingly depicting violence should not be interpreted literally, but rather metaphorically. In other words, "Veg 'n Vlug" doesn't depict an actual event, the artists meant something else entirely, or is perhaps an attempt at sympathetic magic. (To be clear, as far as I know Ferguson has never written about this specific pictograph. I'm illustrating the kind of argument he's made about other rock art depictions of violence). And there are certainly aspects of "Veg 'n Vlug" that isn't literal. Parkington continues from the above quotation:

This bald, but reasonably literal description gives no hint of the intriguing and enigmatic details that impart a deeper, but still obscure meaning to this apparently unified composition. Take the double red lines for example. They cannot, as might appear at first glance, be footprints or a path, because they connect the feet of those in the procession to the bow of one of the cave occupants and emerge from the bowstring to enter the mouth (or face) of the bow and arrow-wielding figure. The strange figure reeling in the lines from the feet of the prone, perhaps dead, figure cannot be manipulating footprints or a path in any literal way. It is likely that the double, parallel red lines are painted to illustrate some connectedness between people that is intangible but central to the meaning of the composition. The attachments to feet, hands, equipment and mouth probably indicate the nature of the connection but are not explicit enough to provide a definitive narrative.

So what does this mean? Well, clearly, the pictograph cannot be strictly literal. Perhaps the artist(s) intended to convey some, now obscure, metaphorical meaning. Perhaps aspects of the drawing represent something abstract. Contra Ferguson et. al., though, I don't think it is reasonable to conclude that a metaphorical interpretation obviates a literal interpretation.

Take my avatar and favorite painting, A Philosopher Lecturing on the Orrery by Joseph Wright:

The painting has two, complementary, meanings. Whether Wright had in mind a specific instance of a scientist^[1] demonstrating an orrery (a clockwork-driven model of the universe), it's clearly the kind of thing that went on at the time. That is, there certainly were orreries, scientists, and scientists demonstrating orreries and the painting represents an instance of the latter. The painting conveys much more than just 'such-and-such' happened, though. Many metaphorical interpretations are possible, naturally, but Wright seems to have intended it as a celebration of science, of the Enlightenment. The point could be argued, but suppose we agree A Philosopher represents the Enlightenment. Does that mean we have to abandon a literal reading? Insist that the painting tells us nothing about orreries and scientists? Obviously not. Literal and metaphorical representations can, and often do, co-exist.

What this illustrates, I'm suggesting, is that however we interpret the non-literal aspects of "Veg 'n Vlug", we need not abandon a literal reading. In other words, even if there are abstract or metaphorical meanings we can assign to the pictograph, it still depicts two groups engaging in violence. It's necessary to go a step or two further, in fact. Not only can metaphorical and literal readings co-exist, we should apply Occam's Razor and favor a kind of interpretive parsimony: the simplest interpretation - the one that requires the fewest new assumptions - is likely the correct one. And in nearly all cases, the literal interpretation is the simplest.

Whatever metaphorical or abstract readings we assign to any pictograph do not necessarily obviate literal interpretations. And interpretive parsimony - favoring the simplest possible interpretation - cautions against metaphorical readings in the first place, and demands especially strong evidence before we elevate metaphor over straightforward representation. In short, unless we have strong reasons to think otherwise, pictographs like "Veg en Vlug" represent evidence of ancient violence.

----

^[1] The word 'philosopher' at the time had multiple meanings, one of which was what we would now call a scientist.

Lazy Linking

"The Durban Boredom Festival"

• So a friend, my fiancée and I went to a local psychic fair recently. I was planning to write about it... but it was a horrid experience, so I never got round it it. Luckily, Angela (the aforementioned fiancée) has written a great account of what went down at the fair and trust me, short as it is, her post contains everything you'll possibly want to know about it. Overall conclusion: way too much incense, rampant woo, boring as hell, complete ripoff.

"The latest verdict on using brain imaging for lie detection"

• BPS Research Digest reports on using fMRI et. al. to spot lying. Short version: it doesn't work. (At least not yet).

"Offensive Play: How different are dogfighting and football?"

• Malcolm Gladwell's latest New Yorker piece in which he compares the morality of dogfighting - almost universally reviled - with that of American football. It turns out that, like with boxing, a football career often results in an Alzheimers-like condition called chronic traumatic encephalopathy. Amazingly, new resarch using accelerometers has revealed players regularly suffer hits to the head of up to 90gs. Not surprisingly this is a Bad Thing that does severe damage to the brain over the long run. Gladwell suggests this may make football morally comparable to dogfighting: the injuries and suffering of the players are an inherent and ineradicable feature of the game.
• As a big rugby fan I couldn't help wondering what the situation is like for my favorite Saturday diversion. Do rugby players also suffer as much damage? Obviously, only research could settle the issue (and some may already exist, I don't know). From the armchair, it's difficult to tell: on the one hand, there are many fewer hits to the head in rugby but, on the other, the players don't wear helmets or much protective gear. My (rather bland) guess, for the little that's worth, is that brain trauma is not as common in rugby as it is in football or boxing, but significantly more prevalent than in the general populace. I'm not going to stop watching though, that's for sure.

"Psychology: A Reality Check" (paywall, I think)

• A great Nature editorial calling for evidence-based clinical psychology in the United States. I'd say it's also much needed elsewhere, the training of psychologists is often criminally devoid of science or even critical thinking. 
• "Clinical psychology at least has its roots in experimentation, but it is drifting away from science. Concerns about cost–benefit issues are growing, especially in the United States. According to a damning report [pdf] published last week an alarmingly high proportion of practitioners consider scientific evidence to be less important than their personal — that is, subjective — clinical experience."
  
• "The irony is that, during the past 20 years, science has made great strides in directions that could support clinical psychology — in neuroimaging, for example, as well as molecular and behavioural genetics, and cognitive neuroscience. Numerous psychological interventions have been proved to be both effective and relatively cheap. Yet many psychologists continue to use unproven therapies that have no clear outcome measures — including, in extreme cases, such highly suspect regimens as 'dolphin-assisted therapy'."

"How We Lost Our Diversity"

• Interesting piece by the excellent Ann Gibbons about new research on the causes of human genetic homogeneity (relative to other primates).
• "Modern humans are a lot alike - at least at the genetic level - compared with other primates. If you compare any two people from far-flung corners of the globe, their genomes will be much more similar than those of any pair of chimpanzees, gorillas, or other apes from different populations. Now, evolutionary geneticists have shown that our ancestors lost much of their genetic diversity in two dramatic bottlenecks that sharply squeezed down the population of modern humans as they moved out of Africa between 60,000 and 50,000 years ago."
• See also: John Hawks' fairly critical analysis of the same study.

"Analogies to apes lead us on the wrong track"

• Razib Khan over at Gene Expression on how Ardi drives home the message that drawing analogies between humans and the other extant apes can be misleading. Six million years is a long time, and there's no reason to think our common ancestor with the chimps and bonobos was particularly chimp-like. Somewhat counterintuitively, the opposite might even be true.

"Dear Penn and Teller: Bullshit!"

• I've only recently remembered that I have Season 6 of Penn & Teller's Bullshit so I'm only watching it now. And like Massimo Pigliucci in the above post, I just hated their episode (6-06) on environmentalism. Libertarians so obviously have blinkers on when it comes to global warming that it positively amazes me that they're not more self-critical. It also reminds us all, of course, that being vigilant about our own biases is important.  

"Islam: A Shifting Focus"

• One of the most widespread misconceptions about Islam is that most of its faithful are Arabs. In actual fact, Asian Muslims vastly outnumber Muslims from other parts of the world, making up 61.9% of the global number of 1.57 billion believers.
• "A new survey of the world’s Muslim population, by the Pew Research Center based in Washington, DC, will help those who are keen to break that link [i.e. the perception that most Muslims are Arabs]. It estimates the total number of Muslims in the world at 1.57 billion, or about 23% of a global population of 6.8 billion. Almost two-thirds of Muslims live in Asia, with Indonesia providing the biggest contingent (203m), followed by Pakistan (174m) and India (160m)."
• "Perhaps more surprising will be the finding that the European country with the highest Muslim population is not France or Germany, but Russia, where 16.5m adherents of Islam make up nearly 12% of the total national population. Compared with other surveys, the report gives a lowish estimate for the number of Muslims in France (3.6m), as it does for the United States (2.5m); in both those countries, secular principles make it impossible to ask religious questions on a census."

Carnival of Evolution 16

• A superb edition of the Carnival of Evolution - there are many worthwhile posts to check out. My pieces on foxes and on chameleons were featured. 

Video: Ardipithecus ramidus

So unless you've been living under a rock for the last two weeks (well, or you don't follow the science news at all) you would have heard about Ardi (a female Ardipithecus ramidus), who is the oldest known hominid and a possible human ancestor. Ardi's remains and her likely habitat was analyzed in detail in a Special Issue in Science and several of the results are very surprising, including that she had arboreal adaptations (i.e. traits for living in trees) despite being bipedal.

Anyway, I don't have much to say (not my field) but I do want to point to this video (embedded below, or click here) that the team at Science produced and that has not received enough playtime. It's a great primer on the significance of the find and what it could tell us about hominid evolution.



(By the way: John Hawks has pointed out that one of the photos used in this video is poorly scaled, so it doesn't give a good indication of the skeletal proportions).

Big new find at Sterkfontein

The Times reports on a big new find at Sterkfontein, ‘The Cradle of Mankind’. Alas, we don’t know what’s been found, as the scientists are keeping it secret until they’re ready to publish. An excerpt:

This much can be revealed: new fossil discoveries have been made by Berger in the Cradle of Humankind. The discovery was disclosed to Parliament a few months ago. President Jacob Zuma recently took a break from his busy schedule to visit Wits to view these new items. So, we know we’re talking about something big. So big, the paleontological world is buzzing with excitement and there is widespread speculation that they will provide new clues to the evolutionary puzzle.

But none of this brings us any closer to answering the question: what precisely has been found? A possible pointer lies in the involvement of Thackeray. The professor has increasingly focused his interest on the field of variability, in size and shape, examining the areas of human evolution where the boundaries start to break down. Modern humans share 98% of their genes with chimpanzees. Studies involving the rate of mutation of DNA have produced a virtual molecular clock, indicating that the chimpanzee/human split occurred somewhere between 5million and 7million years ago. Subsequently, the hominins also split into branches. Several different hominin species have been found at Sterkfontein alone and three major tool cultures have been identified.

Via: John Hawks.

Exploring Ötzi

Discovered in 1991 on a glacier in the Ötztal Alps in Austria, Ötzi the Iceman is a beautifully preserved mummy of a man who died in around 3300 BCE. Amazingly, scientist have pinpointed where he lived (by analyzing the isotopic composition of his teeth), what his last meal was (chamois, red deer and some plants), his possible occupation (copper smelter), among many other details. Importantly, it was also determined that he was a victim of violence: an arrowhead was lodged in his shoulder, there were bruises and cuts to various parts of his body, and he apparently received a blow to the head.

The European Academy Bozen/Bolzano recently released a fantastic photoscan of his entire body, which allows anyone to explore and analyze him in detail. I particularly recommend checking out his tattoos -- interesting stuff indeed.

Top 100 Anthropology Blogs

Rather oddly given that I'm not an anthropologist and this isn't an anthropology blog, Ionian Enchantment has been featured on a list of the Top 100 Anthropology Blogs. Woohoo!
(I think).

John Hakws Weblog, by the way, is a must read if you haven't seen it previously.

Video: Brian Ferguson on war

Rutgers professor Brain Ferguson is one of the leading anthropologists of war and in the Bloggingheads.tv interview embedded below (or click here) he argues the evolutionary state of nature was not Hobbesian (i.e. warlike), but broadly Rousseauian (i.e. peaceful). To be clear, Ferguson does not deny that warfare exists worldwide, his position is that it emerged in the Mesolithic and that the Paleolithic - the period during which humans evolved - was therefore largely peaceful. Ferguson discusses, among other things, Lawrence Keeley's War Before Civilization (which I reviewed here), Azar Gat's War in Human Civilization (which I read recently and will hopefully review soon), Wrangham & Peterson's Demonic Males and Napoleon Chagnon's influential Science paper "Life Histories, Blood Revenge and Warfare in a Tribal Population".

While I don't agree with Ferguson's conclusion, the interview is highly recommended if you are at all interested in prehistory or human evolution.

Quote: Lionel Tiger

I dare say that it remains overwhelmingly the case in the social sciences that almost everywhere it is possible to receive a doctoral degree without studying any other species than humans. Even then, the work is likely to involve people and their behavior in the past generation and in a highly limited geographical area. This is wholly understandable, yet intellectually, it is akin to studying the whole of geology but focusing exclusively on Minnesota or doing botany while ignoring photosynthesis.

-- Lionel Tiger, "The Human Nature Project"

The great divide in anthropology

There is a rather simplistic, but nonetheless interesting, article in the Times Higher Education Supplement on the great divide in anthropology between evolutionary and socio-cultural anthropologists. An excerpt:

On one side are the evolutionary anthropologists. "(They believe) our behaviour is based on things that we did to find mates in our years of evolution," says Alex Bentley, a lecturer in anthropology at Durham University. "Then we have the social anthropologists. Some of them really strongly reject this kind of thinking. They consider it reductionist. They are focused on the specifics of culture."

Put crudely, social anthropologists describe and compare the development of human cultures and societies, while evolutionary anthropologists seek to explain it by reference to our biological evolution. The two sides of the one discipline are struggling to unite.

Do also have a look at the comments, there are some real corkers...

Technology Quarterly

The Economist publishes a fantastic Technology Quarterly that summarizes important recent scientific and technological developments. This quarter's edition has just been released (the other articles are listed in the box at right). Noteworthy articles:

• unusual applications of crowdsourcing,
• how archaeologists are using Google Earth to make discoveries, and
  
• the development of tiny medical robots.

Video: The Genographic Project

Spencer Wells gave a superb talk (embedded below or click here) at the 2007 TED conference about The Genographic Project. The project is a collaboration between National Geographic and IBM to build on the pioneering work of Luigi Luca Cavalli-Sforza and document human genetic variation and chart ancient migration patterns. Very highly recommended.

Books II

Every month or so I write mini-reviews of some the books I have read recently.

Perhaps reading multiple rave reviews set my expectations too high, but I wasn’t blown away Mistakes Were Made (but not by me) by Carol Tavris and Elliot Aronson. To be clear, the book is excellent and thought-provoking, but it’s not quite up there with the recent non-fiction classics (How the Mind Works, Guns, Germs and Steel and so on). Tavris and Aronson are both eminent social psychologists and their aim is to demonstrate just how much people are beset by cognitive illusions such as confirmation bias and cognitive dissonance. The book is at its best when the authors summarize the relevant social psychology demonstrating, for example, that memory is a ‘self-justifying historian’ and fallible in the extreme, that clinical judgments divorced from experimental evidence is a recipe for disaster, that the criminal justice system regularly convicts innocent people because it does not sufficiently account for human fallibility and so on. The book deteriorates towards the end because the authors stray too far from solid evidence: the final chapter (“Letting Go and Owning Up”), for example, is insipid because they don’t seem to have evidence for their thesis that knowing about biases and cognitive dissonance can reduce their effect. Further negatives: Tavris and Aronson continually use a decidedly unhelpful metaphor of “the pyramid” (don’t ask), their emphasis on cognitive dissonance over the other biases seems overdone, some of their examples are a bit pat, and there are some contradictions (they claim, for example, that Americans see mistakes as stupid, but then insist in the final chapter that people who admit making mistakes are honored). Despite these problems, Mistakes Were Made is a fantastic book; the only people who won’t find it enlightening are those already familiar (and comfortable) with the illusions and biases literature. (See also: Tavris’ interview on Point of Inquiry and her blog interview with Greta Christina).

Abusing Science: The Case Against Creationism by Philip Kitcher is an excellent little book that just blows the anti-evolutionists out of the water. Kitcher, a philosopher of science, wrote the book while he was a visiting scholar at the Museum of Comparative Zoology at Harvard University in the company of Stephen Jay Gould, Richard Lewontin and others. At least one good thing and one bad thing resulted: the good thing was this book, which is a scholarly, scientifically well-informed, philosophically savvy and utterly convincing demolition of “scientific creationism”. The bad thing, to speculate somewhat, was Kitcher’s later horrendously misinformed tract against sociobiology, Vaulting Ambition. That is, I strongly suspect Kitcher was convinced by the radical science nutters (led by Gould) that sociobiology was dangerous pseudoscience. Speculation aside, Abusing Science is terrific – it may have been published in 1982, but it’s still relevant for the simple reason that creationism (even its latest incarnation as “Intelligent Design theory”) simply doesn’t change.

I really like Michael Shermer in several ways – I think he’s charming, smart and exceedingly eloquent (there is no better testament to his abilities than his TEDTalk on skepticism). However, I have to admit that I wasn’t all that impressed by Why Darwin Matters: The Case Against Intelligent Design. Firstly, some fairly minor quibbles: Shermer comes across conceited (naming a ‘law’ after himself, talking excessively about his achievements), he seems to be naïve about philosophy of science (in stark contrast to Kitcher) and, inexcusably, I spotted far too many logical fallacies. My real problem with the book, however, is that I thought it was unscholarly and its arguments were mostly unconvincing: I just don’t think Shermer did enough intellectual legwork. (The one exception is Shermer’s admirably clear treatment of the anthropic principle). Obviously, I already agree with his conclusions and I’m certainly not his target audience, but he does not marshal the best arguments for evolution and against intelligent design. That said, I think people who believe in intelligent design or creationism would benefit from Why Darwin Matters and I would recommend it for that audience, those already familiar with evolutionary biology can steer clear. (See also: Shermer’s interview on Point of Inquiry and his talk at the Cato Institute about the book. Note: he comes across far better in person than in the book).

I reviewed No Country For Old Men last time round, and my views of Cormac McCarthy’s latest novel, The Road, is very similar. The novel, very briefly, follows a father and his young son’s journey south as they struggle to survive several years after an unnamed disaster brought an end to civilization and made the cultivation of food impossible. It is a gripping, terrifying and epic tale, told in McCarthy’s inimitable, free-flowing prose. The Road, make no mistake, is a terrific novel and I expect McCarthy to win the Nobel Prize before long. Like with No Country For Old Men, however, I really didn’t like the ending. [Warning: plot spoilers follow]. As you’ll know if you’ve read the book, in the end the father dies of an illness and the boy is rescued by some mysterious man who had been tracking them. I smell a deus ex machina (in the original sense of an arbitrary and convenient plot device introduced to end a story satisfactorily). Every single other character the father and son encounter on their journey is horrible: there are thieves, murderers, cannibals, and blood cultists, among many others. Indeed, even the old man they help turns out to be an ungrateful wrench. Yet, suddenly, just when it is convenient, “the good guys” arrive out of nowhere to save the boy. I hate to sound callous, but McCarthy should have stayed true to his story, stuck to its bleak themes, and let the boy die. [Spoilers end]. That said, read the book, it’s superb. (See also McCarthy interview on Oprah).

Trick or Treatment? Alternative Medicine on Trial by Simon Singh and Edzard Ernst is an outstanding, indeed exemplary, treatment of medical pseudoscience and quackery. Singh, a science journalist, and Ernst, the world’s first professor of alternative medicine, examine four common alternative modalities: acupuncture, homeopathy, chiropractic and herbal medicine. In each case, they relate the treatment’s history, discuss its principles and methods, summarize the relevant clinical evidence and then make an overall assessment. Their conclusions will not surprise skeptics: with the exception of the herb St. John’s Wort (for depression), there is no evidence that any of the above-mentioned modalities have an effect beyond placebo, and some (chiropractic particularly) carry non-trivial risks.
Importantly, Singh and Ernst start the book with a chapter entitled “How Do You Determine the Truth?” which explains the scientific method and, vitally, its rationale, using historical medical advances (James Lind’s cure for scurvy and Florence Nightingale’s innovations, particularly). That first chapter alone is worth the price of the book: it is a perspicacious, detailed, persuasive, and unusually interesting defense of evidence-based medicine. Further virtues: the book is easy to follow but not superficial, the prose is lively and engaging, the explanations and illustrations are clear throughout and, probably most importantly, they get the science exactly right. Pretty much the only negative thing I have to say is that the introduction contains a bit too much foot-stomping and puffery about the authors’ neutrality, love of truth and rigor – a bit of humility is in order. My bottom-line: read this book, buy copies for your quackery-loving friends and family and spread the word. (See also: Singh's interview on The Skeptics Guide to the Universe).

Martin Gardner may have written the first edition of Fads & Fallacies in the Name of Science in 1952 bit it is still relevant and certainly still worth reading. The first thing that stands out about this book is just how dedicated and scholarly Gardner was: he pioneered modern scientific skepticism so he could not rely much on previous work. As a result, he was forced to do an enormous amount of original research – he had to wade through countless obscure pseudoscientific tomes and then rely on his own judgment for a conclusion. Remarkably, despite taking positions on over twenty pseudosciences (sadly, many of which are still current), Gardner got it entirely right in almost all of the details. Moreover, his style is extremely appealing because he avoids hectoring and the temptation to nitpick, being content, on the contrary, to let silliness speak for itself. Finally, Gardner's book is particularly valuable to skeptics because it provides a sense of perspective and proportion: many of our current problems (like celebrities endorsing pseudoscience) is not new, and might have been worse in the past. Fads & Fallacies, in short, is a classic that richly deserves a continued readership.

Lawrence Keeley’s War Before Civilization: The Myth of the Peaceful Savage is an enormously important book that has not received nearly as much attention as it deserves. (By the way, I’ve blogged about an aspect of this book before). Keeley, an archeologist who specializes in Neolithic Europe, argues we have “artificially ‘pacified the past’ and shared a pervasive bias against the possibility of prehistoric warfare” (p. vii). His thesis, in other words, is that prehistoric warfare was at least as violent as modern warfare, and probably much more so. Keeley marshals masses of evidence in support of this conclusion, including excavated fortifications, mass graves, skeletons of unequivocal murder victims, in addition to an argument from ethnographic analogy. The latter argument is especially interesting: to establish the plausibility of prehistoric warfare, he reviews the evidence that current non-state societies are far from peaceful. The evidence here, incidentally, is absolutely overwhelming: over 40% of all people in tribal societies die violent deaths, compared to less than 1% for the U.S. and Europe in the 20th century including both World Wars. Had the world experienced violent death-rates in the 20th century comparable to those of non-state societies, billions of people would have died, instead of hundreds of million. Some negatives: Keeley’s prose is a touch purple in places, he extrapolates a bit too far once or twice (taking findings about trade in non-state societies to make recommendations for state policy), he denigrates Western military strategy too much and I think he caricatures Hobbes’ position. Overall, though, the book is fantastic: it is scholarly, interesting and entirely convincing. A particularly nice quote to give you a feel for the book:

The burned villages, the arrowheads embedded in bones, the death tolls, and the mutilated corpses speak more truthfully, more passionately on this dismal subject than all the recorded verbiage of the living, which is riddled with cant, sophistry, and flights of fancy. The dead voices heard here tell us that war has an ugly sameness; it is always a compound of crimes no matter what kind of society is involved or when in time it occurs. After exploring war before civilization in search of something less terrible than the wars we know, we merely arrive where we started with an all-too-familiar catalog of deaths, rapes, pillages, destruction and terror (p. 173-174).

Rapid human evolution

There is a pretty good article in US News and World Report about the growing evidence for very rapid recent evolution in human beings:

At the same time, the human genome has been scrambling to adapt to a rapidly changing world—11,000 years ago, nobody farmed, nobody milked domesticated animals, and nobody lived in a city. People with a mutation that aided survival were more likely to thrive, reproduce, and pass that mutation along to offspring. For example, the capacity to digest lactose, the sugar in milk, has become common only over the past 3,000 years. Now, about 95 percent of the people in northern Germany have the mutation, which also popped up independently among the Masai in Africa and the Lapps in Finland. Hawks says: "This is really rapid evolution."

Unfortunately, instead of sticking to the science, the second half of the article goes into genetic engineering and its ethical complications. Speculation ensues and the piece ends off with a pathetically pat conclusion. Despite its shortcomings, the article is still a decent popular summary of very important science.

(Via John Hawks, who features in the article).

The Crow Creek Massacre

I just finished reading University of Illinois archaeologist Laurence Keeley's fantastic War Before Civilization: The Myth of the Peaceful Savage. I will write a full review soon, but I wanted to mention an incident Keeley discusses that rather surprised me: the Crow Creek Massacre. Keeley's main argument in the book is that archaeologists have artificially "pacified the past" by overlooking evidence of warfare, murder, cannibalism and genocide in prehistory. Despite my rather Hobbesian view of human nature - which makes it pretty easy for me to accept Keeley's conclusions - it surprised me that such clear evidence of Native American brutality existed. I suppose all those Hollywood films with eco-pacifist Indians had a subconscious effect on me after all.

The Crow Creek massacre occurred around 1325 AD (that is, long before Columbus) in South Dakota, along the Missouri River. The victims were from a group of agriculturalists known as the 'Initial Coalescent people' who are thought to have migrated from the plains due to drought. There is evidence that the victims were malnourished and that several individuals had previous war wounds (healed scalpings, arrows embedded in bone, etc.). Additionally, they clearly felt threatened: a defensive ditch was under construction when the massacre occurred. The total number of victims is estimated at 486 people, which included men, women and children:


Disturbingly, many of the victims were tortured (including "tongue removal, decapitation, and dismemberment"), at least 90% of the people were scalped and their village was burnt. The victims were left to rot where they fell, and were apparently only buried some months later.

This event is quite possibly due to ecological stress but nonetheless serves as a clear existence proof of genocide and abject cruelty among prehistoric societies. Native Americans had clearly tasted evil before Europeans arrived; which is obvious, of course, except that recent intellectual history and popular opinion has denied such a thing was possible. This evidence, though, is incontrovertible.

Homo floresiensis update II

See also my earlier pieces: "The floresiensis mess" and "Homo floresiensis update".

Yet more about whether the Flores specimens discovered in 2004 constitute a new species, this time on the positive side. In an upcoming article in the Proceedings of the National Academy of Sciences, paleoanthropologists Adam Gordon, Lisa Nevell, and Bernard Wood of George Washington University argue a statistical analysis of one of the skulls (LB1) reveals the Flores specimens cannot be shrunken Homo sapiens. Intriguingly, they say the skull most closely resembles Homo habilis, a very primitive hominid indeed. Says Gordon: "This is particularly exciting because ... it suggests that we really do have a hominin lineage that split off from our own as much as 1.7 million years ago, yet persisted up until the time when modern humans started peopling the Americas. That's pretty cool."

When the study goes online, I'll update this entry with a link to it.

Update: the article can be found here.

Homo floresiensis update

The other day I blogged about the continuing debate over Homo floresiensis, the possible species of Homo. I mentioned two recent papers that challenge the proposition that the Flores specimens constitute a new species: Obendorf et. al.'s suggestion that the specimens are Homo sapiens who suffered from cretinism and Berger et. al.'s report on a new find of small-bodied remains on Palau that suggests the Flores specimens might have been Homo sapiens who exhibited physiological dwarfism. The anthropoligist John Hawks has now examined the Berger paper (which he peer-reviewed for PLoS One) and written a detailed blog entry on it. I can, by the way, highly recommend Hawks, he certainly knows what he's talking about.

See also: SciAm Observations Blog's piece on Berger et. al.'s paper and National Geographic's article.

The floresiensis mess

The debate over the status of Homo floresiensis, the possible species of Homo that survived until about 12,000 years ago that was discovered in 2003 on the Indonesian island of Flores, just won't go away. Two separate much-publicized studies that challenge the notion that floresiensisis is a distinct species emerged this week. First, Peter Obendorf, Charles Oxnard, and Ben Kefford argued in their study in Proc. Roy. Soc. B that the purported floresiensis skeletons were actually Homo sapiens who suffered from congenital hypothyroidism, or cretinism. The anthropologist John Hawks, for one, disagreed with this finding on his blog, arguing Obendorf et. al. got it badly wrong.

More recently, University of Witwatersrand paleoanthropologist Lee Berger and colleagues describe a new set of small-bodied human skeletons that were found in Palau, Micronesia. Berger et. al. conclude that their specimens are Homo sapiens who exhibit physiological dwarfism that regularly emerges in island contexts. They also suggest that the characteristics of the Flores specimens "may be best explained as correlates of small body size in an island adaptation, regardless of taxonomic affinity. Under any circumstances the Palauan sample supports at least the possibility that the Flores hominins are simply an island adapted population of H. sapiens, perhaps with some individuals expressing congenital abnormalities."

Rapid human evolution

It is commonly accepted among evolutionary psychologists (at least in the Santa Barbara school*) that "our modern skulls house a stone age mind". As Tooby & Cosmides put it,

natural selection... takes a long time to design a circuit of any complexity. The time it takes to build circuits that are suited to a given environment is so slow it is hard to even imagine -- it's like a stone being sculpted by wind-blown sand. Even relatively simple changes can take tens of thousands of years. The environment that humans -- and, therefore, human minds -- evolved in was very different from our modern environment. Our ancestors spent well over 99% of our species' evolutionary history living in hunter-gatherer societies. That means that our forebearers lived in small, nomadic bands of a few dozen individuals who got all of their food each day by gathering plants or by hunting animals.

Similarly, Edward Hudgens explains,

Evolutionary psychologists downplay the possibility of significant cognitive evolution in the 10,000 or so years since the advent of agriculture (a period of time known as the Holocene) for reasons of both science and political correctness. Scientifically, 10,000 years (500 generations) is not much time for natural selection to act, and it certainly is not enough time to evolve new, complex adaptations—sophisticated mechanisms coded for by numerous genes.

New research just released in PNAS has the potential to undermine these claims fatally. John Hawks and his colleagues argue that human evolution accelerated very rapidly in the last 40,000 years. The abstract:

Genomic surveys in humans identify a large amount of recent positive selection. Using the 3.9M HapMap SNP dataset, we found that selection has accelerated greatly during the last 40,000 years. We tested the null hypothesis that the observed age distribution of recent positively selected linkage blocks is consistent with a constant rate of adaptive substitution during human evolution. We show that a constant rate high enough to explain the number of recently selected variants would predict (1) site heterozygosity at least tenfold lower than is observed in humans, (2) a strong relationship of heterozygosity and local recombination rate, which is not observed in humans, (3) an implausibly high number of adaptive substit utions between humans and chimpanzees, and(4) nearly 100 times the observed number of high-frequency LD blocks. Larger populations generate more new selected mutations, and we show the consistency of the observed data with the historical pattern of human population growth. We consider human demographic growth to be linked with past changes in human cultures and ecologies. Both processes have contributed to the extraordinarily rapid recent genetic evolution of our species.

(See also: John Hawks's two blog entries on his study, and Reuters' report)

*The Santa Barbara school of evolutionary psychology is the best known type of EP, its foremost exponents are Leda Cosmides, John Tooby, Donald Symons, David M. Buss, Steven Pinker, Margo Wilson and Martin Daly. (It's so called because Symons, Tooby & Cosmides are at UC Santa Barbara).