Friday, March 4, 2011

Mate magnet madness: When the range of possible explanations exceeds your own hypothesis

Figure 1. My apologies to Baby Jaguar
for not finding a picture that included
him.
My daughter will be three in just a few weeks. She loves telling stories. These stories have the same, uncomplicated arc every time: she and her friends Dora, Diego, Boots and Baby Jaguar go on an adventure to rescue Mommy from the giant condor. Or sometimes Mommy and Dora and Diego and Boots and Baby Jaguar are rescuing her. Or sometimes Daddy does the rescuing.

There is almost always a net, then a pair of Rescue Scissors needed to cut the captive free. But the variation in these stories is very small, the framework borrowed heavily from one of the few mythologies known to my little girl: Dora the Explorer.

Evolutionary psychology is often a kind of story-telling, and instead of borrowing from a preschool cartoon they borrow from the concept of anisogamy. Anisogamy is sexual reproduction formed by unequal gametes, in our lineage a big egg made by females and little sperm made by males. This provides the foundation for differential reproductive investment, where females often put in the time and effort of gestation, lactation and care. From here, proponents of EP see essential differences between what men and women want in relationships, and the kinds of relationships that are optimal, and a model this broad makes it possible to shoehorn any behavior into its adaptive framework.

Figure 2. The actual image that
accompanied Tierney's column.
Enter John Tierney, my (not) favorite journalist for the New York Times. This is the man who thinks that sexism is a radical act (I am referring to his charming articles on gender disparities in science). So I suppose I shouldn’t have been surprised when he outed himself as an EP fanboi in his most recent piece, “The Threatening Scent of Fertile Women.”

Tierney covered the work of Jon Maner and others who have studied relationship maintenance – the suite of behaviors that keeps a couple together. In particular, Tierney focuses on the problem of the wandering eye, or rather, the possible mechanisms that prevent it in a monogamous couple. The idea here is that relationship maintenance is evolutionarily adaptive, because when a couple stays together it is easier to raise offspring and increase reproductive success.

The range of explanations

The study that frames Tierney’s column is Miller and Maner (2010). Thirty eight undergraduate men rated the attractiveness of a woman with whom they interact, at several points over her menstrual cycle. The authors found NO relationship between where a woman is in her cycle and how attractive a single man finds her, but a negative relationship between the chance a woman is fertile and how attractive a partnered man finds her.

What do Miller and Maner (2010) discuss, and what is the idea Tierney is so enamored with?
“It’s possible that some of the men in Florida were just trying to look virtuous by downgrading the woman’s attractiveness, the way a husband will instantly dismiss any woman pointed out by his wife. (That Victoria’s Secret model? Ugh! A skeleton with silicone.) But Jon Maner, a co-author of the study, says that’s unlikely because the men filled out their answers in private and didn’t expect the ratings to be seen by anyone except the researchers.

“It seems the men were truly trying to ward off any temptation they felt toward the ovulating woman,” said Dr. Maner, who did the work with Saul Miller, a fellow psychologist at Florida State. “They were trying to convince themselves that she was undesirable. I suspect some men really came to believe what they said. Others might still have felt the undercurrent of their forbidden desire, but I bet just voicing their lack of attraction helped them suppress it.””
This conjecture is unconnected to the study’s methodology and results. Nowhere in that study did they assess the participants’ state of mind or ask them how they felt about this. How do we know they were trying to convince themselves of anything? This finding, while interesting, does not test their hypothesis for an evolutionary framework for relationship maintenance that includes adaptively suppressing attraction to others.

Maner et al (2009) studied the attention people pay to images of attractive people of the opposite sex when first exposed to sexual words like “lust” and “kiss.” They recruited 120 straight undergraduates, thirty six of whom were in committed relationships. Individuals in committed relationships paid far less attention to the attractive images than those not in relationships. Tierney titters,
“The subliminal priming with words related to sex apparently activated some unconscious protective mechanism: Tempt me not! I see nothing! I see nothing!
I’ve done my own share of human subjects research, and subjects will often tell you or do what they think you want, or they will just not be honest if they don’t want you to know the truth. What if, as originally posed by Tierney himself, the respondents weren’t warding off temptation but wanted to look virtuous? What if, now bear with me because this might seem crazy, the people in these studies were in love with their partners and genuinely uninterested in anyone else? Too often EP wants to provide a single explanation for a behavior, when the range of possible explanations far exceeds their hypothesis.

An anthropological perspective

Jamie Jones, Associate Professor at Stanford and blogger at Monkey’s Uncle describes anthropology like this,
“…[A]nthropology is the science charged with explaining the origin and maintenance of human diversity in all its forms. To achieve this end, anthropology must be unapologetically grand in its scope. How can we explain human diversity without documenting its full extent, through both time and space, and across cultures? … Where does the tapestry of human diversity come from and how is it that we continually manage to resist powerful homogenizing forces and hang on to our diversity? What commonalities transcend local difference to unite all humanity? How is it that civilizations rise and fall? And what is the fate of humanity?”
Jamie beautifully depicts the importance of documenting and understanding diversity even in the face of efforts to simplify human nature. Thus, to me, an anthropological perspective is often at odds with EP explanations for behavior.

An anthropological perspective asks, what happens if you take these basic observations and, instead of deciding on a favorite explanation and applying it to everyone, put them into a model in which you can vary context (age/sex specific mortality rates, distribution of resources, what have you) and see what range of strategies actually give fitness benefits? That is, when you actually throw some variation into the equation, is this still the best strategy for the partnered men with whom Tierney feels simpatico?

Right now we don’t know. Much psychological empiricism rests on undergraduates who participate in studies for course credit. When one wants to make connections to evolutionary adaptedness, they may be a place to start, but not end.

I have a real problem with continuing to use this population to make statements of universality for all humans. Undergraduates usually are trying to avoid pregnancy and build their financial and social capital, so relationship maintenance for the sake of reproductive success rarely exists. Until we can show that relationship maintenance, and the particular behaviors Miller, Maner and others study within that are shown across many populations, and particularly across reproductively-aged folks, their argument for adaptation fails.

Figure 3. Celebrations of marriage.
Another problem is that most work on relationships in EP tends to be heteronormative, meaning that they think nothing of assuming that either everyone is straight, or the universally best behavioral strategy is to be straight. They also tend to assume that the best strategy is to be monogamous, with occasional sneaky infidelity permitted if one can get better genes or more offspring that way (keep in mind that there is a difference between what might be biologically advantageous in a certain context, and what is culturally appropriate – the argument here is not against the culture of monogamy).

But heterosexual monogamy is only one reproductive strategy of many that humans employ. Depending on how you measure it, monogamy and polygyny (single male, multi female marriage) vie for the most frequent strategy – in fact, polygyny occurs in about 80% of modern human societies (Murdock and White 1969). There are even a few rare populations that practice polyandry, which is the marriage of a single female and multiple males. And, even in those populations where monogamy is practiced, serial monogamy is far more frequent than lifetime monogamy: this means that individuals have a series of monogamous relationships rather than find one mate for life (so no, divorce is not a modern human invention).

When taking an even broader, comparative perspective, monogamy isn’t practiced by our closest relatives at all. Chimpanzees and bonobos, both equally related to us, are promiscuous. This is a scientific term for a reproductive strategy that involves females and males making reproductive decisions to mate with many individuals at each fertile period. Bonobos are also promiscuous, but they also use heterosexual and homosexual sex to reduce stress and aggression, and form bonds among one another. Gorillas, our next closest relative, are polygynous. Orangutans are very solitary, but essentially promiscuous. It’s only once you delve into the lesser apes, the gibbons, that you see any monogamy, and they are far less monogamous than we first thought (Brockelman et al 1998).

Maintaining a heterosexual, monogamous relationship is certainly advantageous at certain times, in certain contexts. But it is not universally adaptive, even within humans. Without anyone studying these behaviors in populations that use different reproductive strategies, and in the absence of comparative data to support these assertions, we are at an impasse.

Conclusion

In the words of a friend, EP is plugged into evolutionary theory with little more than a ratty old extension cord. EP takes some very basic, ancestral conditions, like differential costs of reproduction, and uses it in a sufficiently vague way that any behavior can relate to females generally being the ones to put in all the time and effort into making babies. Yet EP often ignores the three conditions necessary for natural selection, the mechanism for evolution. For natural selection to act on a trait, the trait must be variable, heritable, and produce differential reproductive success. Rarely does EP understand variation in a trait, rarely does it examine whether said trait has a genetic component, and rarely does it test whether their trait confers a reproductive advantage.

Are fertile women a threat to partnered harmony, their scents providing a temptation that noble men must suppress? I can’t rule it out, but I also think it is one of the least likely of many possible explanations.

Unfortunately for readers of the New York Times, Tierney loved this idea more than he loved interrogating it.

Acknowledgements

I’d like to thank Charles Roseman, friend, faculty curmudgeon and Bastard Colleague from Hell, for taking a look at an early draft of this post and providing commentary crucial to its improvement. Any rhetorical or scientific errors are my own.

References

Brockelman, W., Reichard, U., Treesucon, U., & Raemaekers, J. (1998). Dispersal, pair formation and social structure in gibbons ( Hylobates lar ) Behavioral Ecology and Sociobiology, 42 (5), 329-339 DOI: 10.1007/s002650050445

MANER, J., GAILLIOT, M., & MILLER, S. (2009). The implicit cognition of relationship maintenance: Inattention to attractive alternatives Journal of Experimental Social Psychology, 45 (1), 174-179 DOI: 10.1016/j.jesp.2008.08.002

Miller, S, & Maner, J (2010). Evolution and relationship maintenance: Fertility cues lead committed men to devalue relationship alternatives Journal of Experimental Social Psychology, 46, 1081-1084

Murdock, G., & White, D. (1969). Standard Cross-Cultural Sample Ethnology, 8 (4) DOI: 10.2307/3772907

Image sources

Dora picture: http://www.doratheexplorertvshow.com/dora/dora-explora-pics.htm
Lady magnet: http://www.nytimes.com/2011/02/22/science/22tier.html?_r=2&ref=johntierney
Same-sex marriage: http://markusisthedrug.onsugar.com/date/2009/05/07

22 comments:

  1. Hear, hear! I read that Tierney article and made a mental note to figure out why it annoyed me so much. Glad to see thoughtful criticism.

    ReplyDelete
  2. Thanks Christopher and roese, glad it was useful!

    ReplyDelete
  3. Very nice treatment, Dr. Clancy, and a particularly clear unpeeling of how the differential-reproductive-effort argument can lead to slippery EP arguments. I knew this study got a lot of attention. Given that, I was a bit astonished, reading this, to learn that (atop the other problems) the study used only 38 men, and all of them undergrads. I've not read the study so hesitate to chastise the authors for overextending results from such a limited sample. But I'm dismayed -- shocked but not surprised, an all-too-familiar feeling — that so many sci writers did.

    Of course we want to know what makes people tick, and our interest is understandably quickened when the ticking relates to sex. But really.

    ReplyDelete
  4. Thanks, David :).

    I only cited two of the four studies Tierney discusses, but I read them all, and they all study undergrads. I also have a number of statistical quibbles with how they ran the data (there were unequal sample sizes in some, but they didn't run nonparametric tests, which would have been more honest but usually reduces the significance of one's results). I left all that whinging out because I was trying to stay focused on my anthro-related issues. But yes, the actual study grates quite a bit.

    ReplyDelete
  5. Great post! Very nice analysis and taking-apart of some shaky premises. Like commenter roese above, I agree that something seemed suspicious about this but this makes it clear. With regard to your very nice "I have a real problem with continuing to use this population [undergraduates] to make statements of universality for all humans," readers may want to check out the article "The weirdest people in the world?" by Henrich, Heine and Norenzayan, Behavioral and Brain Sciences 2010 33(2-3).

    ReplyDelete
  6. I also wonder how John Tierney thinks the college-aged guys in the study *knew* which of the women were ovulating so as to avoid them! He seems to take it as a matter of course that young men are able to do that, but I can't even tell when I'm ovulating, much less whether anybody else is.

    ReplyDelete
  7. I appreciate the dismantling of Tierney’s poorly constructed/conducted study (not to mention the blind assumptions that were made – gah! – which ultimately biased their conclusions), but as someone with an armchair interest in the field of EP, it seems as though a stance is taken against the entire subject... or am I missing something? Are the fields of EP and Anth diametrically opposed?

    ReplyDelete
  8. I really enjoyed reading this essay, thank you for your work here. The article by Tierney was so aggravating, and Miller and Maner's study was so unconvincing. Their study seems to represent a new brazen tactic of post-hoc pseudo-theorizing: when faced with negative results, propose that the subjects were "suppressing" the urges that would have generated the correlation you were fishing for. Please.

    ReplyDelete
  9. Jason - yes, that is a wonderful article, and it hits on one of the big issues I have with EP. I really should write a post on that some time. Thanks for reminding me of some interesting work.

    Lindsay - I agree! What I thought was interesting was that they hit on the sexy "pheromone" idea in the title ("threatening scent") but didn't follow up with it in the column. Of course, it's a pretty hotly contested idea, because some studies have suggested that we can detect pheromones, where others have shown the opposite. I'm kinda in the "we may secrete them but probably can't detect them" camp. But I could be convinced the other way with the right experimental design and study.

    gkaylor - maybe what I should say is that I am pretty opposed to EP, particularly the EP that makes the news, and I think a lot about anthropology is opposed to EP... but I know plenty of anthropologists who are trying to reach out to EP folks in order to try and figure out a more nuanced way to study behavior. There is such a thing as a middle ground between the two, but few tread it.

    Brian - thanks! I like your term "post-hoc pseudo-theorizing" -- I think it reflects the majority of the quotes Tierney pulled from the study authors for his column. It was pretty inexcusable stuff, complete storytelling unrelated to the experimental design of the study they actually performed.

    ReplyDelete
  10. I enjoyed this post Professor Clancy, I wish I tuned into your blog sooner. I find this article very interesting. Your analysis starting with your daughter then connecting it to the study sparked my interest. This is the first time I am hearing of the study done, however I have to disagree with even the thought behind the experiment. Men must have been the ones who came up with the idea because any woman knows they do not feel their prettiest during menstruation. Thus they may not come off as attractive to some men, but in the same light what one man likes another may not, so unless the studied group was a group of males with the same preferences and taste in women, the study couldn't possibly turn out usable results.

    ReplyDelete
  11. Jamie Sukowicz (249)March 8, 2011 at 11:15 AM

    I really like the points you made about this study. I completely agree that undergraduate males are by no means an accurate representation of the human population at large. Plus, using only 38 is not at all sufficient. I do think that it would have been helpful to ask these subjects why they felt the way that they did, and it is possible that not all of the subjects were heterosexual. Additionally, they may not have perceived a woman as attractive due to where she is in her cycle—they may just find her more attractive, for a variety of different reasons completely unrelated to reproduction.

    ReplyDelete
  12. This blog really had me questioning my point of view on the matter when it mentioned gay marriages. It seemed to me that the existence of gay marriage seems to disprove much of the EP theory. I am very liberal in my opinions on same sex marriages, I believe that if someone loves a person it doesn't matter which sex the person is. The EP theory made a lot of sense to me though, it seemed biologically beneficial to have emotional attachment to a single person increasing the probability of reproductive success. I think what I liked most about this article was that I didn't simply agree with it but am now more confused than when I started reading it and will continue to question my own opinion on both subjects. I just wonder if they are mutually exclusive or if I'm "allowed" to feel that both sides are "correct."

    ReplyDelete
  13. Jenessa Conner (249)March 8, 2011 at 10:29 PM

    A very interesting perspective! I think its important to note that surveys are very tricky and as Jamie said, 38 undergraduate males is hardly a accurate representation. Just a good example of how studies should be examined very carefully, without jumping to any quick conclusions. The concluding marks about how the EP theory is difficult to link to the three characteristics of natural selection is a topic worthy of debate. I would enjoy to here an counter-argument as why Pro-EPers such as Anna feel that natural selection is lacking in the EP theory.

    ReplyDelete
  14. Sherry Xiao (249)March 9, 2011 at 12:56 AM

    I think it is unfair on how women are looked at when they are in their menstrual cycle by men. Women are the ones that are keeping the reproduction system going on and having offsprings in this world. From watching today's video of "The Business of Being Born," women go through so much when they are giving birth. It is something that is so empowering in their lives and unforgettable. Undergraduate men do not understand the meaning of the birth process and so I also think that there could have been better subjects used instead. Men who has seen the birth process might be better subjects since they see the pain that women has gone through.

    ReplyDelete
  15. Anthony Garcia (249March 9, 2011 at 1:08 AM

    A very interesting article for sure. First, I found the part about undergraduate test subjects pretty true. I have partaken in experiments as a test subject, and I probably was not the best subject. Students are not exactly focused on the experiment. Also, I have heard the theory about how are most recent relatives are not monogamous in Anthropology 143, and find that to be a great argument for people who say that marriage is a "natural" thing. Lastly, this theory does seem like a pretty general idea with very little proof from an experiment.

    ReplyDelete
  16. I suspect when you write about EP conclusions in terms of "put them into a model in which you can vary context (age/sex specific mortality rates, distribution of resources, what have you) and see what range of strategies actually give fitness benefits?" they would simply respond with their second-favorite concept of "facultative adaptation"--the simple idea that one can make beneficial adjustments to the phenotype based on external cues from the environment. Just like tanning when exposed to more UV radiation, it is possible to make behavioral/psychological adjustments to mate preferences based on who is available. This would be their standard answer.

    But your suggestion about throwing in some variation and looking at the range of fitness benefits can also run into trouble, since selection will change the mean/variances of a distribution of phenotypes (that causally coavary with a phenotype) and if this distribution is flat or uniform, then there will be little or no selection...this is fine and good, but then we can't use the mechanics of evolutionary theory to investigate human mating preferences. Shuster and Wade, in their 2003 book, discuss this issue pertaining to everyone "making the best of a bad job." If everyone is behaving optimally, we either have a uniform fitness distribution or we will still have some doing better than others, and selection will start to operate. Of course humans can cognitively adjust to circumstances over the short-term and these ephemeral actions might escape the scrutiny of selection; what is particularly interesting is that for a field that calls itself "evolutionary psychology" there is huge emphasis on the former and only a little lip service paid to the latter. EP is much more E than P.

    Just a note, I don't think that most work in EP assumes monogamy is the best strategy. I haven't seen that stated in any of the main textbooks (Buss argues for serial-monogamy and fairly rampant male infidelity). Usually, it is the opposite. If anything, monogamy has been recruited in attacks of EP research, since EP dogma trots out the "men are promiscuous and prone to abandon their mate" stuff. Monogamy, and more recently, communal rearing of offspring (e.g., Hrdy's latest book) have been the response to the ostensible male proclivities to abandon their mate. David Buller brought up this idea in his fabulous critique of EP "Adapting Minds".

    And you're absolutely right: extrapolating results from undergrads to the whole world is about as dumb as shoe-horning 7 billion unique genotypes/personalities into 12 categories (i.e., Astrology)

    ReplyDelete
  17. Anth 249 students, thanks for your comments!

    Rich, thanks for commenting. I think you're right about the monogamy thing, I probably overstated my case as it relates to EP as a whole. However, the recent work on relationship maintenance, particularly the work cited here, emphasizes monogamy for the sake of child-rearing and discusses selection pressures favoring parents staying together. The implication, or at least what I'm taking out of the papers, is that monogamy is the best strategy for increasing reproductive success, given the high investment humans tend to make in their offspring. They don't discuss serial monogamy (which I agree is more Buss's position, for instance) and if anything, serial monogamy would operate against their premise, since it leads to a fair bit of step-parenting.

    I also like your point about EP being more E than P... though again, the E is framed in a way that assumes adaptation underlies all behavior, which I doubt is the position that most behavioral ecologists would hold. In fact, as I'm thinking about it, I wonder if I would be nearly as annoyed with EP if they would just reframe things to acknowledge variability in strategies. And, you know, study someone other than undergrads once in a while.

    ReplyDelete
  18. Bateman's and Trivers' treatment of anisogamy is not diminished because all male-female reproductive tactics and strategies fail to conform to the baseline model explaining the consequences of differential energy investment in gametes. Deviations from baseline provide opportunities to study each of these cases in order to unpack the causes of these patterns.

    ReplyDelete
  19. Misaki Suehiro (249)March 9, 2011 at 6:42 PM

    I have to agree with the rest of the 249-ers on questioning the universality of the Tierney's statement, especially due to such a small sample size. I couldn’t help feeling that there were some over-generalized statements when relating the data to the female population as I read through Tierney’s article. Also, I like Anthony’s comment about taking into account this idea of a “natural” view of relationships, looking at this in relation to our animal relatives, as well culturally-- especially keeping in mind the changing view of homosexual marriage in our society as a more widely accepted (in some cases, at least).

    ReplyDelete
  20. Laura Carroll (249)March 9, 2011 at 8:34 PM

    I definitely agree as well in regards to the test subjects. Undergraduate students are not the best people to be polling when trying to gather data. Personally I know when I have had to partake in experiments for my psychology class, I was more focused on getting it over with than actually being involved in the experiment. But on a more important note, I found the fact that the various species of monkeys rarely have monogamous relationships absolutely astounding! It makes you wonder if monogamy is more a product of our cultural/religious practices than actual instincts? Maybe our instinct to cheat/ look lustfully at other men/women is really not so bad at all? I mean maybe we are just doing what our bodies are telling us to. Seems like quite a bit of cognitive dissonance to me. Overall this is an incredibly interesting post Professor Clancy! I can't wait to read more!

    ReplyDelete
  21. It appears that John Tierney got a little excited at plugging in some simplistic research into a complex issue. I have to say that the critique is very well structured and very convincing. It just sounds like a classic blinders view as far as trying to make a certain theory fit. One thing that I have learned for sure was that there is a lot of variation concerning evolutionary research. Another thing that I have learned outside of the classroom from first hand experience is that a monogamous long term relationship is not at all simple. When you hear theories about all the mechanisms against monogamy as a practice in humans, it is easy to jump to conclusions just as John did with EP, so I understand why he might be so quick to try and simplify things. That being said I think when you try to simplify such a complex idea as relationships, especially within the context of evolution and reproductive success, a lot of mistakes can be made and incorrect assumptions concluded. Monogamy may be a difficult and under researched area in human behavior but that is no reason to over simplify or condemn it as a futile reproductive effort.

    ReplyDelete