Keeping It Topical: That Catcalling Video

Viral fame is an interesting thing. It can come out of nowhere and disappear just as quickly; not unlike a firework. It can also be rather difficult to predict, due to the fact that eventual popularity can often be determined largely by preexisting popularity. This week, one such story that appears to have been caught up in a popularity spiral has been the subject of catcalling: specifically, a video of a woman documenting around 100 instances of unsolicited comments she accumulated while wandering the streets of New York City for 10 hours (which is about one such comment each 6 minutes). At time of writing, the video has around 33 million views, not counting the various clone videos (which is around 6 million such views a day, making for such pleasant numerical symmetry). Unsurprisingly, there’s been a lot of talk about the video; a pile which I’m about to add to. Perhaps the most common conversations have been had concerning whether it’s appropriate to call some of the unsolicited comments the woman received “harassment” (for example, “Have a nice evening”, said in passing, or the various comments suggesting she is “beautiful”).

  Can’t a girl be dating a guy for two years and not get bombarded with harassing proposals?

On that front, there are some natural barriers in perspective that might make consensus hard to reach, owing to what these propositions are thought to represent: solicitations for causal sex. Men, for instance, would likely find such solicitations or comments generally pleasant when receiving them from women, whereas women tend to have precisely the opposite reaction (Clark & Hatfield, 1989). Given the perceptual flavor that such comments often have, men might tend to see them as less of a big deal than women (though sex is hardly all there is too it; such an effect would also be influenced by one’s mating strategy – whether they prefer long- or short-term sexual relationships – as well as other such interacting variables). A second barrier to consensus on the matter is the concentrated nature of such comments: whereas the woman in the video might have received over 100 comments that she views as annoying, they are also coming from over 100 different men. If individual comments aren’t viewed as a problem, but an aggregate of them is (kind of like pollution), discussions over whether they should be condemned might hit some snags in attempting to reach agreement.

A second discussion that has been had about the video concerns the racial component. In the viral video, the majority of the men on the street making these comments are non-white. Subsequent analysis of the video led to the conclusion that around 60% of the comments in question were received on a single street in Harlem. Whether this location was specifically selected in order to solicit more comments, whether certain comments from other people in other areas were edited out, or whether the comments were simply received primarily from the people in that area are unknown, but it does leave a lot to be desired in terms of research methods. It’s important to bear in mind that this video was not a research project for the sake of gathering new information: it was a video designed to go viral that ends with a donation link. Any video which failed to generate appropriate reactions from people on the street would be unlikely to be used, as I can’t imagine video of someone walking around the street without incident encourages people to empty their wallets effectively.

In the interests of furthering that discussion, it’s also worth considering a reported cross-cultural replication attempt of this study. Psychological research has often been criticized for relying on WEIRD samples, and reliance on a single person (with an agenda) from largely a single street should not be taken to be representative of people’s experiences more generally (either in that city or aboard). So, when a woman in New Zealand apparently tried the same thing – wandering the streets of a city for, I presume, 10 hours – it’s worth noting that the video reports her receiving a total of two comments, one of which was a man asking for directions. Assuming the walking time was the same, that’s the difference between a comment every 6 minutes and a comment, with different content, every 300. As seems to be the case in psychology research, flashy, attention-grabbing results don’t always replicate, leaving one wondering what caused the initial set of results to be generated in the first place. Statistical variance? Experimental demand characteristics? Improper sampling?

  Divine intervention, perhaps?

It’s difficult to say precisely what caused the difference in men’s behavior between videos, as well as why most of the comments were made in one specific area in the first one. The default answer most people would likely fall back on would, I imagine, be “cultural differences”, but that answer is sufficiently vague to not actually be one. This is the part where I need to be disappointing and say that I don’t actually have an answer to the questions. However, I would like to begin some speculation as to the psychology underlying the sending of these unsolicited comments and, from there, we might be able to figure out some variables which are doing some of the proverbial lifting here.

One possibility is that these comments are used by men specifically to intimidate women, or make them feel otherwise uncomfortable and unwelcome. As some might suggest, these comments are just an extension of a male culture that hates women because they’re women and will take about every chance it gets to ruin their day (variants of this hypothesis abound). I find such an explanation implausible for a number of reasons, chief among which is that calling someone beautiful is unlikely to be the most effective way of expressing contempt for them. When black people in America were marching for civil rights, they were not met with protesters telling them to “have a great day” or admiring their bodies with a suggestive “damn”. Such an explanation likely mistakes an outcome of an event for its motivating cause: because some women feel uncomfortable with these comments, some people think the comments are made in order to make women uncomfortable. This conclusion is likely the result of people wishing to condemn such comments and, in order to do so, they paint the perpetrators in the worst possible light.

However, it’s worth noting that, as far as I can tell, there are some women who either (a) express flattery at these comments or (b) sadness that they are not the targets of such comments, taking the lack of comments to say something negative about their attractiveness (which might not be inaccurate). While such sentiments may or may not be in the minority (I have no formal data speaking to the issue), they paint a much different picture of the matter. Typically, people experiencing violence, oppression, and/or hatred, do not, I think, need to be assured that they aren’t actually being complimented; the two are quite easy to tell apart most of the time. In fact, in the original video, at least one of the men is explicit about the notion that he is complimenting the girl (though admittedly he does go about it in a less than desirable fashion), while another man asks whether the reason the woman isn’t talking to him is that he’s ugly. If, as these ancedotes might suggest, catcalling is tied to factors like whether she is attractive or unattractive, or the response to it tied to the man’s desirability, it would be difficult to tie these factors in with misogyny or intentional harassment more generally.

“Why does my friend always get the harassment? Is it my hair?”

There is, of course, also the other end of this issue: men getting catcalled. While, again, I have no data on the issue, the misogyny explanation would be hard to reconcile with gay men or women making such comments towards men (even if such comments are likely less common owing to the historical costs and benefits of short-term sexual encounters for each sex). The simpler explanation would seem to be that such unsolicited comments, while not necessarily desired by the recipient, are earnest – if clumsy – attempts to start conversations or lead to a sexual encounter. Given that similar comments tend to be made in first messages on dating websites, this alternative seems reasonable (women who complain about receiving too many one line messages online should see the parallels immediately). The problem with such attempts is unlikely to be with any particular one being deplorable so much as it is their sheer volume.

Now it is quite unlikely that these comments ever yield successful encounters, as I mentioned above. This could be one reason they are often considered to be something other than friendly or sexual in nature (i.e., “since this behavior rarely results in sex, it can’t be about getting sex”; the same kind of error I mentioned earlier). The rarity of sexual encounters resulting from them is also likely why the proportion of men making them is really very low even though they’re rather cheap – in terms of time and energy – to make. While 100 comments in 10 hours might seem like a lot, one also needs to consider how many men the woman in question passed in that time, in one of the largest cities in the world, who said nothing. For every comment there were likely several dozen (or hundred) men who made no attempt to talk to the actress. Any explanation for these comments, then, would need to pinpoint some differences between those who do and do not make them; general aspersions against an entire gender or culture won’t do when it comes to predictive accuracy. For what it’s worth, I think a healthy portion of that variance will be accounted for by one’s sexual strategy, one’s current relationship status, the attractiveness of the person in question, and whether the target is sending any signals correlated with sexual receptivity.

What predictions can be drawn from alternative perspectives I leave up to you.

He’s Got Your Eyes…Right?

Last post, I was discussing paternal investment in children. The point of that post was to draw attention to the fact there are often rather good biological reasons for why we might expect men and women to be differentially interested in investing their time and energy in raising children versus doing other adaptive things. This is not to say that we shouldn’t expect men to be interested in investing in children, of course; just that we shouldn’t expect such things to be indiscriminate or motivated by the same factors as women’s altruism. I wanted to expand on one of those ideas a bit more today: specifically, the idea that men lack assurance in their paternity when fertilization takes places inside the female, whereas women can be 100% certain the child they give birth to is theirs. “Certain”, in the former context, refers to the notion that women were unlikely in need of a solution the adaptive problem of maternity certainty, as giving birth to a child was an honest and reliable signal that the child was related to the mother genetically.

“Little does she know it’s not her child” – No one, ever

The first study I wanted to draw attention to concerns the resemblance of a child to their parents. Naturally, as children inherit half of their genes from each parent, we should expect that children tend to resemble their parents with respect to a number of external and internal features. That much is pretty noncontroversial. However, since fathers cannot be assured of their paternity, we might expect men to attend to certain similarities between them and their would-be children when calculating the likelihood that a given child is actually theirs. If an Asian woman gives birth to a Black child, her Asian husband can likely be fairly assured that the child in question is not, in fact, his, and there might have been some infidelity involved somewhere along the line. One might – and, indeed, some have – make the corresponding argument that we might expect children to physically resemble their father more than their mother. The logic would go roughly as follows: if the child resembles their father more, they might receive more parental investment from the father, as he can be more certain the child is his. So, since that tends to be a good thing for the survival and reproductive prospects of the child, we might expect children to bias their resemblance towards their fathers (foregoing for the moment the precise mechanisms through which that might be achieved).

There’s a major issue that such expectations run into, however: reality. In a 1999 paper, Berdart & French collected photos of parents and their children from 28 families. The children’s photographs were taken when they were approximately 1, 3, and 5 years old; the parent’s pictures were all from when the child was about 1 year old. During each trial, the participants (180 undergraduates) were presented with a single picture of a child alongside three women or three men and asked to try and identify the child’s parent. This process was repeated 28 times for each subject. If children tend to resemble their fathers more than their mothers, we should also expect people to be better at matching the child to their father than their mother. This effect failed to materialize, however: for one-year old children, the average correct matching to the father (13/28) was not different than correct matching to the mother (12/28); similar findings obtained for the three-year olds (13/28 and 13/28, respectively) and five-year olds (14/28 and 14/28, respectively).

So while child did certainly tend to resemble their parents consistently (though clearly less than perfectly), they failed to resemble their fathers more than their mothers. Berdart & French (1999) suggested that there might be a rationale for this lack of distinct resemblance: if fathers were good at figuring out which children were theirs, they would presumably also be good at figuring out which children were not theirs, and withhold investment from the latter group. I don’t want to spend too much time on this point other than to note that it’s not a particularly strong one, as it would be good for the ostensible fathers if they had such a skill, and the ill effects on the unrelated child shouldn’t be expected to have an impact on its adaptive nature. Nevertheless, the important point here is that children do not appear to actually resemble their father anymore than their mother. Reality need not always get in the way of perceptions, though.

Fitness is 99% instagram filters

In general, having perceptions that match up to reality is a good thing. If you think you can succeed where you will fail, you’re more likely to waste your time; it you think you will fail when you can succeed, you’ll miss opportunities. Things like that. The exception to that rule concerns contexts of persuasion. There are beliefs I might prefer you to have because they would make me better off, rather than because they are true. So, if some part of my brain that deals with persuading others holds an incorrect belief, that’s not necessarily a problem. As the last post touched on, if people can convince others certain sex differences are due to sexism, that might have some useful implications for certain groups of people to the extent that people are trying to avoid being perceived as sexist and are willing to take steps to remediate the situation. For the present purposes, however, if women are trying to extract investment from men for their children, it would be in those women’s best interest if the man in question believes the child in question is actually related to him, as men are more likely to invest in child on that basis alone. Accordingly, we might predict that women will be more likely to try and convince men that their children resemble them.

Enter some research by Daly & Wilson (1982) that examined the spontaneous utterances of people following the birth of a child. Their first sample consisted of 111 births which had been taped; the fathers were present in 42 of them. From those 111 tapes, approximately 70 comments about the baby’s appearance were recorded. When it was the mother speaking, she remarked on the baby’s resemblance to the father in 16 instances, and the baby’s resemblance to herself 4 times. By contrast, when the father was speaking, he remarked on how the baby resembled the mother 4 times, and himself only once. Further, every time the resemblance to the mother was commented on, the utterance was singular; when the father’s resemblance was being discussed by the mother, however, six of them contained repetitions (e.g. “He looks like you…He’s got your eyes”). Immediately following the birth of a child, then, the resemblance to the father appeared to focused on specifically by the mother. While the mother was nominally more likely to comment on the resemblance to the father in his presence (75%) relative to his absence (47%), this difference didn’t reach significant with the small sample size.

A follow-up study surveyed the responses of mothers, fathers, and the relatives of both concerning the child’s resemblance. Responses came back from about 230 parents and 150 relatives. In all cases, each group suggested the child looks more like the father than the mother by a ratio of at least 2:1. This is in slight opposition to the previous results insomuch as both mothers and fathers said the child looked more like the father. This may have something to do with sampling bias, though, as only about 1/5 of the sample returned any surveys. It seems plausible, as the authors note, that “...fathers rankled by any serious suspicion of nonpaternity would be unlikely to find the questionnaire an amusing diversion“. It possible, then, that fathers might be overstating their physical resemblance to the child in surveys as signals of their unwillingness to abandon investment in the child or relationship, but that’s just speculation on my part. The videos, by contrast, might have proven to be more of an unbiased sample, freer from demand characteristics. Though it’s difficult to say, it’s worth noting that around 25% of the survey respondents reported that “everyone” in their life said the baby looked like the father, as compared with no one reporting comparable utterances about the mother. This is in spite of the finding that children don’t particularly resemble the father over the mother in a matching task, suggesting that such comments might represent social politeness, rather than accurate perceptions.

“She looks just like both of you…”

To repeat the major point here, there can be benefits to perceiving the world in inaccurate ways when you are trying to convince other people of things. Whether that thing is the resemblance of the child to a parent or whether sex differences are due sexism is quite irrelevant. It is likely, in many of these cases, that the part of the brain doing the talking legitimately believes those perceptions so as to better convince others, while different parts of the brain might disagree. Now, in this case, we happen to have data to suggest that the perceptions – or at least what people say about their perceptions – are incorrect; we also have a relatively straightforward theory for explaining why we might expect this might be the case. In many other cases we are not so fortunate.

References: Berdart, S. & French, R. (1999). Do babies resemble their fathers more than their mothers? A failure to replicate Chistenfeld & Hill (1995). Evolution & Human Behavior, 20, 129-135.

Daly, M. & Wilson, M. (1982). Whom are newborn babies said to resemble? Ethology & Sociobiology, 3, 69-78.

Understanding Male Investment In Children

As a rather social species, humans seem to have a fairly advanced theory of mind. By that, I mean we attribute things like beliefs, intentions, desires, and so on to other people in efforts to understand, explain, predict, and manipulate the behavior of others. The adaptive value of this skill can be hard to overstate and, accordingly, we ought to expect people to be pretty accurate at figuring out the mental states of others. That said, doing so with perfect accuracy is not an easy task, despite our general proficiency with it. Part of the reason, of course, is that things like beliefs and desires are not themselves directly observable, requiring us to make certain assumptions about the reasons for the observable behavior of others. Another part of the reason, however, is that people often have a vested interest in convincing others about certain internal states of affairs, and that interest persists even in the absence of truth value. For instance, if my suffering tends to draw investment from others in the form of social or material resources, it might pay for some cognitive mechanism of mine to over-represent how much I am suffering publicly to others.

“No, really; I am in that much pain. Just come a little closer and see…”

As an example of the trickier aspects of figuring out the intentions and motivations of others, I wanted to use a case of paternal investment in humans. In many mammalian species, males do not tend to assist in the raising of offspring at all. This is owing largely to the fact that males cannot be assured of their paternity the way females can “know” the child they give birth to is theirs. Human males, by stark contrast, often offer substantial investment in children. However it came about, males in our species managed to largely solve the adaptive problem of paternity uncertainty. The key word in that last sentence, though, is largely: we still can’t be sure that a child is ours 100%, so we might expect that, in general, men are less interested in investing in children than women tend to be, especially if the specter to infidelity has been raised. We might also expect that outcome to obtain owing to opportunity costs; what else we could be doing with the time spent investing in children. Time and energy that I spend investing in raising a child is often time and energy I can’t spend doing other adaptive things, like pursuing additional mating opportunities. As the obligate costs to reproduction are lower for men than women, we might also expect men are more interested in putting their time into pursuing mating opportunities and less interested in putting into investment in children, relative to women.

Now these are theoretically-sound evolutionary reasons for expecting the sex with less obligate investment and genetic certainty (typically males) to be less interested in parenting efforts. The logic of managing these adaptive problems should be instantiated in the psychologies of men and women, and to the extent that men and women face different problems, we should perceive the world and behave in different ways. However, some people don’t like the idea that there is any difference between men and women with respect to how interested they are in raising children. As an example, I would offer this article over at Patheos calling Sam Harris a sexist for suggesting men and women have some different interests when it comes to raising children. In it, the author puts forth two hypotheses: that women being the ones to disproportionately cut their careers short to raise children is due to either “…biology or sexism“. In this case, the author favors the “sexism” explanation which, I think, is that men and women and psychologically indistinguishable with respect to their interest in raising children, and would be just as likely to do so were it not for whatever culture is setting different standards. The author’s theory of mind, then, says that men are just as interested in raising children as women (or at least that’s what one part of her mind says publicly). The notable quote I would consider to outline this hypothesis is, “There is no biological reason men and women cannot share the responsibility of childrearing“. No biological reason making such an outcome impossible, sure; just that a host of them make an equitable distribution of interest in doing so fairly unlikely.

The piece and subsequent comments sections are full of anecdotes about how people know men who are supremely nurturing towards children. I don’t doubt that’s the case, just like I don’t doubt that there are some women who are taller than most men. Variance is a thing, after all, and males in our species do tend to invest in children. It doesn’t follow, though, that there are no aggregate differences in desire rear children between the sexes owing to more than sexist culture. The important thing worth noting here is that desires to invest in children are being inferred from the behavior of investing. The problem with doing so is that people might enact behavior for reasons other than desiring to enact the behavior itself. An easy example is a man visiting a prostitute: just because the man gives the prostitute money, it does not mean his motives are altruistic; he is giving her money instrumentally. If he didn’t have to give her money for the sex, he probably wouldn’t. That sounds simple, I’m sure, but how about the hypothesis that men invest in child rearing for reasons owing to mating effort, rather than parenting effort?

“HA; Got one! Bring on the ladies!”

This brings me to a paper by Anderson, Kaplan, & Lancaster (1999). Now this study doesn’t speak to the matter of sex differences in interest in children, as it only focuses on male behavior, but it makes the point well that inferring motives from behavior can be a problem. The study examined male investment in children in conjunction with their relationship status with the child’s mother. There were four groups the men were placed into: (1) those who had genetic children and were in a relationship with the mother, (2) those who had genetic children and were not in a relationship with the mother, (3) those who had stepchildren with a woman they were in a relationship with, and (4) those who had stepchildren from a past relationship. The researchers had a sample of approximately 1,300 men with offspring in one or more of those categories. The question at hand was whether or not a male’s investment in said children varied as a function of the male’s relationship status with the child’s mother and the children themselves. The male’s investment was considered in four groups: (1) Time spent with the child, (2) money spent on the child, (3) whether the child attended college, and (4) how much support the child received for college.

I want to focus on the monetary investment category, as I feel it’s the easiest to think about, (and because all four classes of investment showed basically the same pattern). With respect to monetary investments over the past year to children 17 or younger, genetic children from a previous relationship received, on average, about $700 less than similarly-aged genetic children from the current relationship (less, in this case, refers to how much a statistical model accounting for a number of factors predicted the typical child should get). By contrast, stepchildren from the current relationship received only around $150 less than genetic children from a current relationship.Stepchildren from a previous relationship received about $1,500 less than genetic children from the current relationship, and $900 less than genetic children from previous relationships. For children age 18-24, the same pattern held, with the exception of the genetic children from past marriages receiving more money than stepchildren from the current one, though the two categories did not differ significantly.

These results found that men do indeed tend to invest in children; often substantial amounts. This fact was never in question. However, the amount they invested in the child, whether in terms of time or money, varied contingent on their genetic relatedness to the child and relationship with the child’s mother. Some of men’s interest in investing in children, like women’s, owes to their relatedness to the child: genetic children from past relationships received much more investment than stepchildren from past relationship. This is classic kin selection. I presume very few people would suggest that parents tend to invest in their own children more because “their culture tells them to do so”, rather than positing some kind of biologically-grounded reason. It also seems like a hefty portion of the investment in children by men could reflect mating effort towards the mother: the men behaved as if they were trying to build or maintain a relationship with a woman through investing in her children. Sure; it might not be as romantic as a dinner date, but investment is investment. It follows that men might well be less interested in raising children per se, but quite interested in maintaining a relationship with the mother, so they invest at certain levels despite their lack of intrinsic interest. Put another way, it is quite plausible that women with children do not generally wish to be in relationships with partners that abuse or neglect the child, so men try to avoid that in order to not be ruled out as mates.

“Ladies…”

Now, again, I don’t have comparable data for women, but the point at hand is that just because you find men investing in children, it doesn’t mean that their sole motivation is in the investing per se. We could very well find that men and women invested relatively equally (or unequally) in children and that their motives for doing so differ substantially. It is also possible that the people agreeing with the sentiments expressed in the Patheos article represent something of a biased sample, insomuch as they don’t know many men who dislike taking care of children because they wouldn’t want to (and purposefully don’t) associate with such men in the first place. At the very least, I doubt any of them are giving fathers who ran out on their children pats on the back and telling them they understand. Finally, it is also possible that people might be inferring certain motivations on the part of one sex or the other in hopes of convincing people of some particular political viewpoint or to affect a change in their behavior. Though I don’t have much time to speculate about it, if people have a vested interest in seeing sexism as being responsible for a difference between men and women, you can bet they will find it. Similarly, psychological researchers often have a vested interest in finding certain statistical results and, lo and behold, they tend to find them too. If you’d like to speculate more about men and women’s interest in raising children, sexist biases, and the like, I’ll leave you with some helpful places to do just that.

References: Anderson, K., Kaplan, H., & Lancaster, J. (1999). Paternal care by genetic fathers and stepfathers I: Reports from Albuquerque men. Evolution & Human Behavior, 20, 405-431.

“There Are No Girls On The Internet”

“I’ve discovered through the internet you can do anything you want so long as no one sees your face; it’s like the wild west over here” -Carl

Today is another leisurely day for me, so I’ll be writing about something less research based and more in the realm of argumentative fun. Many people have recently become aware of the site 4chan, owing to the site being the platform for the recent massive leak of celebrity nude photos acquired from breaches of their accounts on iCloud servers. The leak has been dubbed “The Fappening”, which seamlessly combines the internet’s collective love of both masturbation and M. Night Shyamalan puns. In any case, as anyone remotely familiar with 4chan should know, the users, at least some and perhaps most of them, pride themselves on the fact that the site is widely considered to be a cesspool of the internet’s waste. This allows them a certain leisure in expressing views which are, shall we say, less than orthodox. There is a saying originating from the site that goes, “There are no girls on the internet”, though most of you have probably heard it by another name: “Tits or gtfo”. Examining this phrase in somewhat greater detail provides us with an interesting window in men and women’s psychology: both in terms of how we tend to perceive the world, and how others in the world tend to perceive and react to us in turn. Buckle up, because today should be fun.

Always take proper precautions when venturing into the internet

So let’s start with a quick breakdown of the phrase, “There are no girls on the internet”. One 4chan user helpfully provides the meaning of the phrase here, and the heart of the idea is as follows: in offline life, people tend to respond to women in certain, positive ways simply because they are women, rather than because of anything else particularly noteworthy about them. By contrast, the user implies that life on the internet is more of a meritocracy where gender should play no particular role in how people respond to you. Accordingly, when women try to draw attention to their gender online, they are trying to cheat the system and receive a certain type of preferential treatment on that basis alone; the implication is that people online don’t, or shouldn’t, take kindly to that kind of behavior. This idea of, “there are no girls on the internet” was then morphed into the phrase “tits or gtfo”, with the latter phrase suggesting that if women want to call attention to their gender, they should just post a naked picture of themselves as an admission that there is nothing else interesting about them and they can’t stand on their own personality and intellect without relying on their gender to support them.

Now this sentiment might strike some people as profoundly misogynistic, perhaps owing to the manner in which it is expressed. At it’s core, though, it seems to be a rather egalitarian idea: gender shouldn’t matter when it comes to how people interact with each other and preferential treatment on that basis should be done away with. The reason I’m discussing this sentiment is to contrast it with another perspective I’ve come across recently; one that suggests women aren’t welcome on the internet. This perspective holds that women online – and offline, for that matter – are subject to disproportionate amounts of harassment simply because they are women, rather than owing to any kind of behavior they enact or things they say. These two perspectives seem to be at substantial odds with one another with respect to one critical detail: do people like women for being women, or do people hate women for being women?

Obviously, the question is too simplistic and paints the issues with far too broad of a brush to be a meaningful one, but let’s try to answer it anyway; just for fun. To answer such a question one needs to begin with some kind of standard as for what counts as appropriate or inappropriate treatment. Let’s return to the Fappening as an example. Some posts on Jezebel.com find it appalling that certain sites won’t take down the nude pictures of Jennifer Lawrence, citing concerns for the privacy and sensitivity of the women in question as the justification for their being removed. Other posts suggest that it is good that charitable donations motivated by the Fappening are being refused, because the money isn’t coming from the right places. Now that we know Jezebel’s stance on the matter of respecting people’s privacy, we can turn to their sister site, Gawker.com (both sites are owned by Gawker Media). Gawker seemed more than happy to take a stand against respecting people’s privacy by previously posting links to the Hulk Hogan sex tape, suggesting that “we love to watch famous people have sex” and are not terribly troubled by the fact that Hogan was secretly filmed and did not want this video to be released; they were so unimpressed by Hogan’s complaints, in fact, that they tried to refuse to comply with his request to have it removed.

Sure, the situations are a bit different: Jennifer had her privacy breached by people breaking into her online account, whereas Hogan was covertly and unknowingly filmed. While I can’t say for certain whether the writers at Jezebel would be totally happy with someone filming himself having sex with Jennifer without her knowledge and releasing the tape despite her protests, my inclination is to think they would condemn such actions. Also, to the best of my knowledge, no one has claimed that whoever released the Hulk Hogan tape “loathes men” and “wants to punish men just for existing”, though some have suggested this as being the motivation for the Fappening pictures being stolen, just substituting “women” for “men”.

“The only conceivable reason to want to see her naked is because you hate women…”

Though not conclusive by any means, these two cases suggest that it’s plausible that the same behavior directed at, or enacted by, men and women will not always be met with a uniform response. Maddox, over at The Best Page In The Universe, recently put out a new article and video outlining other instances of this kind of double stand with respect to comic book characters; a topic which I have touched on before myself with respect to both superheroes and Rolling Stone covers. In the video, Maddox shows, quite clearly, that Spiderman and Spiderwoman have been depicted in almost identical poses on the cover of comics, but the female version was apparently perceived to be overly sexualized and an embarrassment by some, despite the male version apparently never being noticed. There’s also the research I’ve covered before suggesting that women appear to get reduced sentences for similar crimes, relative to men, if you’re looking for something less anecdotal.

Now none of this is intended to generate some kind of competition over whether men or women, as groups, have it worse. Rather, the point of this analysis is to suggest that men and women, on the whole, tend to have it differently. There are relatively-unique adaptive problems that each sex has tended to face over our evolutionary history, and, as such, we should expect some differences in the psychological modules possessed by men and women. This can cause something of a problem when it comes to discourse regarding whether, say, women are facing a disproportionate amount of harassment online, because what counts as harassment in the first place might be perceived differently; we are all swimming in seas of subjective perceptions that our minds create, rather than bringing them in from the outside world in some kind of objective fashion. What is “threatening” to one person might be innocuous to another, depending on the precise nature of the stimulus and of the mind perceiving it.

For instance, Amanda Hess references an uncited study that found “accounts with feminine usernames incurred an average of 100 sexually explicit or threatening messages a day. Masculine names received 3.7.” Why are “sexually explicit” and “threatening” messages grouped together in that sentence? Well, if the results of Clark and Hatfield’s classic 1989 study are any indication, it’s likely because women might perceive a good deal of unsolicited sexual attention as unpleasant or harassment. However, men might receive that same sexual attention as pleasant and welcome. It is also likely that women will receive a lot more unsolicited sexual attention from men than men will from women, owing the minimum requisite biological costs to reproduction. Grouping “threatening” or “harassment” in the same category as “propositioning” strikes me as precisely the type of thing that can lead to disagreement over how much harassment is going on online. (I think this study is what Amanda is referring to, in which case “feeling horny?” counts as a threatening or sexually explicit message; it’s certainly one of those things, anyway…)

This is a somewhat long-winded way of suggesting that men and women might, and likely do, tend to both perceive the world differently and expect to be treated in particular fashions. If people expect some standard of treatment they are not receiving, they might come to perceive the treatment they get as being overly hostile, unwelcoming, or unfair even if they receive the same treatment as everyone else. This point works just as well for people reacting to the treatment of others: if I expect you to get a certain level of treatment and you don’t, I might try to come to your aid and condemn others for how they behaved on your behalf. That’s not to say that people are, in fact, getting equal treatment in all cases regardless of gender (they often don’t); just to point out that our perceptions of it might differ even if they were.

I’m not saying that such treatment isn’t hostile either; plenty of treatment people receive online is downright threatening, from death threats, to abuse fantasies, to plain old public shaming and ridicule. I’ve received a series of what one might consider abusive messages from strangers online after winning a game we were playing, and that was only after 30 seconds to five minutes of interacting without even talking in a recreational activity; an experience not unique to me by any means. One could imagine that the frequency and intensity of this abuse increases substantially as one becomes more publicly known or begins voicing controversial opinions widely (like calling an entire subculture bigoted or not supporting dedicated servers for your FPS).

“Thanks for your thoughtful message, XxXx420NoScopeFgtxXxX”

In fact, one very reasonable suggestion is that the vitriol present in some of the harassment people receive online is designed specifically to get a rise out of the person receiving it; it’s the M.O. of the internet troll. When it comes to women receiving harassment, for instance, we might expect that women receive particular types of abuse because women tend to be most bothered by it, but they do not receive abuse because they are women. The goal of those sending the abuse is not to make some kind of social or political statement about an issue or express contempt for an entire gender; it’s just to get under someone’s skin.  However, when a different group is being targeted for harassment, the content of the harassment should be expected to shift accordingly.

A good example of this would be 4chan’s trolling of the MMA fighter War Machine (which, I am told, is now his legal name): when users on 4chan found out that War Machine’s father had died after his son’s unsuccessful CPR attempt, they began to tell War Machine he had killed his father (on the anniversary of the death, I would add). This harassment didn’t take that form because people hate those who perform CPR, fathers, MMA fighters, or men more generally; it only took that form because it was what people thought would get the best rise out of him. Judging from the subsequent self-inflicted injuries War Machine documented publicly, the attempt was pretty successful.

“That’ll show ‘em…”

However, just like the immediate point of many trolling comments is to upset others, rather than to make some honest statement, the reactions people have to online harassment should be expected to be every bit as strategic as the trolls themselves, even if not consciously so. Just like the Gawker sites don’t appear to be consistently concerned with privacy (“Yes” with respect to Jennifer, “No” with respect to Hogan), and just like people don’t perceive Spiderman and Spiderwoman to be equally sexualized despite near identical poses on their covers, so too might outrage over online harassment not be evenly spread between targets, even if the harassment itself is quite similar. So, whether the internet is a place of general equality with respect to gender or hostility towards women depends, in no small part, on what kind of treatment people are expecting each gender to receive.

That said I wouldn’t want to accuse any person or group of over-reacting to the harassment they receive just for being them; I’m sure that harassment is particularly unique, and evidence of a widespread bias against you and your friends.

Some Free Consulting Advice For Improving Online Dating

I find many aspects of life today to be pretty neat, owing largely to the wide array of fun and functional gadgets we have at our disposal. While easy to lose sight of and take for granted, the radical improvements made to information technology over my lifetime have been astounding. For instance, I now carry around a powerful computer in my pocket that is user-friendly, capable of accessing more information than I could feasibly process in my entire lifetime, and it also allows me to communicate instantly with strangers and friends all over the globe; truly amazing stuff. Of course, being the particular species that we are, such technology was almost instantly recognized and adopted as an efficient way of sending and receiving naked pictures, as well as trying to initiate new sexual or dating relationships. While the former goal has been achieved with a rousing success, the latter appears to still pose a few more complications, as evidenced by plenty of people complaining about online dating, but not about the ease by which they can send or receive naked pictures. As I’ve been turning my eye towards the job market these days, I decided it would be fun to try and focus on a more “applied” problem: specifically, how might online dating sites – like Tinder and OkCupid – be improved for their users?

Since the insertable portion of the internet market has been covered, I’ll stick to the psychological one.

The first question to consider is the matter of what problems people face when it comes to online dating: knowing what problems people are facing is obviously important if you want to make forward progress. Given that we are species in which females tend to provide the brunt of the obligate parental investment, we can say that, in general, men and women will tend to face some different problems when it comes to online dating; problems which mirror those faced in non-internet dating. In general, men are the more sexually-eager sex: accordingly, men tend to face the problem of drawing and retaining female interest, while women face the problem of selecting mates from among their choices. In terms of the everyday realities of online dating, this translates into women receiving incredible amounts of undesirable male attention, while men waste similar amounts of time making passes that are unlikely to pan out.

To get a sense for the problems women face, all one has to do is make an online dating profile as a woman. There have been a number of such attempts that have been documented, and the results are often the same: before the profile has even been filled out, it attracts dozens of men within the first few minutes of its existence. Perhaps unsurprisingly, the quality of messages that the profiles receive can also liberally be considered less-than optimal. While I have no data on the matter, almost every women who has talked with me about their online dating experiences tends to remark, at some point, that they are rather afraid of meeting up with anyone from the site owing to a fear of being murdered by them. Now it is possible that such dangers are, for whatever reason, being overestimated by women, but it also seems likely that women’s experiences with the men on the site might be driving some of that fear. After all, many of those same women also tell me that they start off replying to all or most of the messages they receive when they set up a profile, only to quickly stop doing that owing to their sheer volume or unappealing content. There are also reports of men becoming verbally aggressive when turned down by women, so it seems likely some of these fears about meeting someone from the site are not entirely without merit (to be clear, I think women are probably no more likely to be murdered by anyone they meet online relative to in person; it’s just that strangers online might be more likely to initiate contact than in person).

The problems that men face are a bit harder to appreciate for a number of reasons, one of which is likely owing to the fact that they take longer to appreciate. As I mentioned, women’s profiles attract attention within minutes of their creation; a rather dramatic effect. By contrast, were one to make a profile as a man, not much of anything would happen: you would be unlikely to receive messages or visitors for days, weeks, or months if you didn’t actively try to initiate such contact yourself. If you did try to initiate contact, you’d also find that most of it is not reciprocated and, of the replies you did receive, many would burn out before progressing into any real conversation. If men seen a bit overeager for contact and angry when it ceases, this might owe itself in part to the rarity with which such contact occurs. While being ignored might seem like a better problem to have than receiving a lot of unwanted attention (as the latter might involve aggression, whereas the former does not), one needs to bear in mind that without any attention there is no dating life. Women might be able to pull some desirable men from the pool of interested ones, even if most are undesirable; a man without an interested pool has no potential to draw from at all. Neither is necessarily better or worse than the other; they’re just different.

Relationships: Can’t live with them; can’t live without them

That said, bickering about whose problems are worse doesn’t actually solve any of them, so I don’t want to get mired in that debate. Instead, we want to ask, how do we devise a possible resolution to both sets of problems at the same time? At first glance, these problems might see opposed to one another: men want more attention and women want less of it. How could we make both sides relatively better off than they were before? My suggestion for a potential remedy is to make messages substantially harder to send. There are two ways I envision this might enacted: on the one hand, the number of messages a user could send (that were not replies to existing messages) could be limited  to a certain number in a given time period (say, for instance, people could send 5 or 10 initiate messages per week). Alternatively, people could set up a series of multiple choice screening questions on their profile, and only those people who answered enough questions “correctly” (i.e. the answer the user specifies) would be allowed to send a message to the user. Since these aren’t mutually exclusive, both could be implemented; perhaps the former as a mandatory restriction and the latter as an optional one.

Now, at first glance, these solutions might seem geared towards improving women’s experiences with online dating at the expense of men, since men are the ones sending most of the messages. If men aren’t allowed to send enough messages, how could they possibly garner attention, given that so many messages ultimately fail to capture any? The answer to that question comes in two parts, but it largely involves considering why so many messages don’t get responses. First, as it stands now, messaging is largely a costless endeavor. It can take someone all of 5 to 60 seconds to craft an opening message and send it, depending on how specific the sender wants to get with it. With such a low cost and a potentially high payoff (dates and/or sex), men are incentivized to send a great many of these messages. The problem is that every man is similarly incentivized. While it might be good for any man to send more messages out, when too many of them do it, women get buried beneath an avalanche of them. Since these messages are costless to send, they don’t necessarily carry any honest information about the man’s interest, so women might just start ignoring them altogether. There are, after all, non-negligible search costs for women to dig through and respond to all these messages – as evidenced by the many reports from women of their of starting out replying to all of them but quickly abandoning that idea – so the high volume of messages might actually make women less likely to respond in general, rather than more.

Indeed, judging by their profiles, many women pick up on this, explicitly stating that they won’t reply to messages that are little more than a “hey” or “what’s up?”. If messaging was restricted in some rather costly way, it would require men to be more judicious about both who they send the message to and the content of those messages; if you only have a certain number of opportunities, it’s best to not blow them, and that involves messaging people you’re more likely to be successful with in and in a less superficial way. So women, broadly speaking, would benefit by receiving a smaller number of higher-quality messages from men who are proportionately more interested in them. Since the messages are not longer costless to send, that a man chose to send his to that particular woman has some signal value; if the message was more personalized, the signal value increases. By contrast, men would, again, broadly speaking, benefit by lowering the odds of their messages being buried beneath a tidal wave of other messages from other men, and would need to send proportionately fewer of them to receive responses. In other words, the relative level of competition for mates might remain constant, but the absolute level of competition might fall.

Or, phrased as a metaphor: no one is responding to all that mess, so it’s better to not make it in the first place

Now, it should go without saying that this change, however implemented, would be a far cry from fixing all the interactions on dating sites: some people are less attractive than others, have dismal personalities, demand too much, and so on. Some women would continue to receive too many unwanted messages and some men would continue to be all but nonexistent as far as women were concerned. There would also undoubtedly be some potential missed connections. However, it’s important to bear in mind that all that happens already, and this solution might actually reduce the incidence of it. By everyone being willing to suffer a small cost (or the site administrators implementing them), they could avoid proportionately larger ones. Further, if dating sites became more user-friendly, they could also begin to attract new users and retain existing ones, improving the overall dating pool available. If women are less afraid of being murdered on dates, they might be more likely to go on them; if women receive fewer messages, they might be more inclined to respond to them. As I see it, this is a relatively cheap idea to implement and seems to have a great deal of theoretical plausibility to it. The specifics of the plan would need to be fleshed out more extensively and it’s plausibility tested empirically, but I think it’s a good starting point.

In The World Of The Blind, The Woman With A Low WHR Is Queen

If you happen to have memories of watching TV in the 90s, chances are you might remember the old advertisements they used to run for the Cinnamon Toast Crunch cereal. The general premise of the ads followed the same formula: “Person X is really good at seeing Y, but can they see why kids love the taste of Cinnamon Toast Crunch?”. Inevitably, the answer was always “no”, as adults are apparently so square that they couldn’t wrap their minds around the idea that children happen to like sugar. Difficult concept, I know. Now, obviously, adults aren’t nearly so clueless in reality. In fact, as you’re about to see, even adults who are really rather poor at seeing things can still “see”, so to speak, why men tend to find certain features in women attractive.

Sauron majored in sociology, so he guessed “cultural conditioning”

The first paper up for consideration is a 2010 piece by Karremans et al. The researchers begin by noting that men appear to demonstrate a preference for women with relatively-low waist-to-hip ratios (WHRs). Women with low WHRs tend to have figures that resemble the classic hourglass shape. Low WHRs are thought to be found attractive by men because they are cues to a woman’s fertility status: specifically, women with lower WHRs – around a 0.7 – tend to be more fertile than their more tubular-shaped peers. That said, this preference – just like any of our preferences – does not magically appear in our minds; every preference needs to develop over our lives, and development requires particular input conditions. If these developmental input conditions aren’t met, then the preference should not be expected to form. Simple enough. The question of interest, then, is what precisely these conditions are; what factors are responsible for men finding low WHRs attractive?

One ostensibly obvious condition for the development of a preference for low WHRs might be visual input. After all, if men couldn’t see women’s WHRs – and all those unrealistic expectations of female body type set by the nefarious media – it might seem awfully difficult to develop a taste for them. This poses something of an empirical hurdle to test, as most men have the ability to see, Thankfully for psychological research – though not so thankfully for the subjects of that research – some men, for whatever reasons, happen to have been born blind. If visual input was a key condition for the development of preferences for low WHRs in women, then these blind men should not be expected to show it. While large samples of congenitally blind men are not the easiest to come by, Karremans et al (2010) managed to recruit around 20 of them.

These blind men were presented with two female mannequins wearing tight-fitting dresses. One of these mannequins had a WHR of 0.7 – around what most people rate as the most attractive – and the other had a slightly-higher 0.84. The blind men were asked to feel and rate each mannequins on attractiveness from 1 to 10. Additionally, the researchers recruited about 40 sighted men to complete the task as well: 20 completing it while blindfolded and 20 without the blindfold. Of note is that all data collection was carried out in a van (read: “mobile laboratory room”) because sometimes psychological research is just fun like that.

“How about coming into my van to feel my mannequins?”

The first set of results to consider come from the sighted men, who completed the task with the full use of their eyes: they gave the mannequin with the low WHR a rating of around an 8, whereas the mannequin with the higher WHR received only around a 6.5, as one might expect. In the blindfold condition, this difference was reduced somewhat (with ratings of 7.5 and around 7, respectively), suggesting that visual input might play some role in determining this preference. However, visual input was clearly not necessary: the blind men rated the low WHR mannequin at around a 7, but the high WHR mannequin at about a 6. In the words of the fine people over at Cinnamon Toast Crunch: “even blind men who can’t see much of anything can still see why men love the figures of women with low WHRs”.

Further evidence from earlier research points towards a similar conclusion (that these preferences are unlikely to be the result of portrayals of women in the media). A paper by Singh (1993) analyzed a trove of data on the female bodies that appeared in Playboy as centerfolds (from 1955-1965 and 1976-1990) and that won Miss America pageants (1923-1987). One might imagine that depictions of women in the media or found to be attractive might change somewhat over six decades if the type of women being portrayed were favored for some arbitrary set of reasons. Indeed, there was one noticeable trend: the centerfolds and pageant winners tended to be getting a little bit skinnier over that time period. Despite these changes in overall BMI, however, the WHR of the groups didn’t vary. Both grounds hovered around a consistent 0.7. Presumably, if blind men were consuming pornography, they would prefer the women depicted in Playboy just as much as non-blind men do.

“We’ll be needing more “databases” for the mobile laboratory room…”

Given the correlation between WHR and fertility, this consistency in men’s preferences should be expected. That’s not to say, of course, that these preferences for low WHR aren’t modifiable. As I mentioned before, every preference needs to develop, and to the extent that certain modifications of that preference would be adaptive in different contexts, we should expect it to fluctuate accordingly. Now that matter of precisely what input conditions are responsible for the development of this preference remain shrouded: while visual inputs don’t seem to be necessary, the matter of which cues are – as well as why they are – are questions that have yet to be answered. For what it’s worth, I would recommend turning research away from the idea that the media is responsible for just about everything, but that’s just me.

References: Karremans, J., Frankenhuis, W., & Arons S. (2010). Blind men prefer a low waist-to-hip ratio. Evolution and Human Behavior, 31, 182-186.

Singh, D. (1993). Adaptive significance of female physical attractiveness: Role of waist to hip ratio. Journal of Personality and Social Psychology, 65, 293-307.

The Best Mate Money Can Buy

There’s a proud tradition in psychological research that involves asking people about how much they value this thing or that one, be it in a supermarket or, for our present purposes, in a sexual partner. Now there’s nothing intrinsically wrong with doing this kind of research, but while there are certain benefits to it, the method does have its shortcomings. One easy way to grasp a potential issue with this methodology is to consider the dating website Okcupid.com. When users create a profile on this site, they are given a standard list of questions to answer in order to tell other people about themselves. Some of these questions deal with matters like, “What are six things you couldn’t do without?” or “what are you looking for in partner?”. The typical sorts of answers you might find to questions like these are highlighted in a video I really like called “The Truth About Being Single“:

“All these people keep interrupting my loneliness!”

The problem with questions like these is that – when they are posed in isolation – their interpretation can get a bit difficult; they often seem to blur the lines between what people require and what they just want. More precisely, the ratings people give to various items or traits in terms of their importance might not accurately capture their degree of actual importance. A quick example concerns cell phones and oxygen. If you were to ask people on Okcupid about five things they couldn’t do without on a day-to-day basis, more people would probably list their phones than the air they breathe. They would also tell you that, in any given year, they likely spend much more money on cell phones than air. Despite this, air is clearly the more important item, as cell phones stop being useful when the owner has long since asphyxiated (even if the cell phone would allow you to go out playing whatever bird-themed game is currently trending).

Perhaps that all seems very mundane, though: “Yes, of course,” you might say, “air is more important than iPhones, but putting ‘I need air’ on your dating profile or asking people how important is the air they breathe on a survey doesn’t tell you much about the person, whereas iPhone ownership makes you a more attractive, cool, and intelligent individual”. While it’s true that “people rate breathing as very important” will probably not land you any good publications or hot dates, when we start thinking about the relative importance of the various traits people look for in a partner, we can end up finding out some pretty interesting things. Specifically, we can begin to uncover what each sex views as necessities and what they view as luxuries in potential partners. The key to this method involves constraining the mate choices people can make: when people can’t have it all, what they opt to have first (i.e. people want air before iPhones if they don’t have either) tells us – to some extent – where their priorities lie.

Enter a paper by Li et al (2002). The authors note that previous studies on mating and partner selection have found sex differences in the importance placed on certain characteristics: men tend to value physical attractiveness in a partner more than women, and women tend to value financial prospects more than men. However, the ratings of these characteristics are not often found to be of paramount importance, relative to ratings of other characteristics like kindness, creativity, or a sense of humor (on which the sexes tend to agree). But perhaps the method used to derive those ratings is missing part of the larger picture, as it was in our air/iPhone example. Without asking people to make tradeoffs between these characteristics, researchers might be, as Li et al put it, “[putting the participants in] the position of someone answering a question about how to spend imaginary lottery winnings”. When people have the ability to buy anything, they will spend proportionately more money on luxuries, relative to necessities. Similarly, when people are asked about what they want in a mate, they might play up the importance of luxuries, rather than necessities if they are just thinking about the traits in general terms.

“I’m spending it all on cans of beans!”

What Li et al (2002) did in the first experiment, then, was to provide 78 participants with a list of 10 characteristics that are often rated as important in a long-term partner. The subjects were told to, essentially, Frankenstein themselves a marriage partner from that list. Their potential partners would start out in the bottom percentile for each of those traits. What this means is that, if we consider the trait of kindness, their partner would be less kind than everyone else in the population. However, people could raise the percentile score of their partner in any domain by 10% by spending a point from their “mating budget” (so if one point was invested in kindness, their partner would now be less kind than 90% of people; if two points were spent, the partner is now less kind than 80% of people, and so on). The twist is that people were only given a limited budget. With 10 traits and 10 percentiles per trait, people would need 100 points to make a partner high in everything. The first budget people started with was 20 points, which requires some tough calls to be made.

So what do people look for in a partner first? That depends, in part, on whether you’re a man or a woman. Women tended to spend the most – about 20% of their initial budget (or 4 points) – on intelligence; men spent comparably in that domain as well, with about 16% of their budget going towards brains. The next thing women tended to buy was good financial prospects, spending another 17% beefing up their partner’s yearly income. Men, on the other hand, seemed relatively unconcerned with their partner’s salary, spending only 3% of their initial budget on a woman’s income. What men seemed much more interested in was getting physical attractiveness, spending about 21% of their initial budget there; about twice what the women spent. The most vital characteristics in a long-term partner, then, seemed to be intelligence and money for women, and attractiveness and intelligence for men, in that order.

However, as people’s mating budget was increased, from 20 points to 60 points, these sex differences disappeared. Both men and women began to spend comparably as their budgets were increased and tradeoffs became less pressing. In other words, once people had the necessities for a relationship, they bought the same kinds of luxuries. These results were replicated in a slightly-modified second study using 178 undergraduates and five traits instead of ten. In the final study, participants were given a number of potential dates to screen for acceptability. These mates were said to be have been rated along the previous 5 characteristics in a high/medium/low fashion. Participants could reveal the hidden ratings of the potential dates for free, but were asked to reveal as few as possible in order to make a decision. As one might expect, men tended to reveal how physically attractive the potential mate was first more than any other trait (43% of the time, relative to women’s 16%), whereas women tended to first reveal how much social status the men had (35% of the time, relative to men’s 16%). Men seem to value good looks and women tend to value access to resources. Stereotype accuracy confirmed.

A now onto the next research project…

This is the reason I liked the initial video so much. The first portion of the video reflects the typical sentiments that people often express when it comes to what they want in a partner (“I just want someone [who gets me/to spend time with/ to sleep next to/ etc]“). These are, however, often expressions of luxuries, rather than necessities. Much like the air we breathe, the presence of the necessities in a potential mate are, more or less, taken for granted – at least until they’re absent, that is. So while traits like “creativity” might make an already-attractive partner more attractive, being incredibly creative will likely suddenly count for quite a bit less if you’re poor and/or unattractive, depending on who you’re trying to impress. I’ll leave the final word on the matter to one of my favorite comedians, John Mulaney, as I think he expresses the point well: “Sometimes I’ll be talking to someone and I’ll be like, “yeah, I’ve been really lonely lately”, and they’ll be like, “well we should hang out!” and I’m like, “no; that’s not what I meant”.

References: Li, N., Bailey, J., Kenrick, D., & Linsenmeier, J. (2002). The necessities and luxuries of mate preferences: Testing the tradeoffs. Journal of Personality and Social Psychology, 82, 947-955.

Gender Gaps Vs. Gender Facts

In a now-classic 1994 paper, John Tooby and Leda Cosmides discussed an idea they labeled “instinct blindness”. One of the basic ideas of this paper is that our minds have evolved to become really good at doing particular things; so good, in fact, that we often don’t realize we’re doing them. Vision, for instance, is an incredibly complex problem to solve. Our cognitive systems for vision function so automatically, however, that we don’t realize the depth of the problems inherent in seeing; we simply open our eyes and see, typically without any conscious experience of the task being particularly challenging. A consequence of this instinct blindness is that much of our psychological functioning goes, essentially, unnoticed; in other cases, facets of our psychology are viewed as not needing an explanation (i.e. “It’s just natural that…”) because they just seem so normal. Now instinct blindness doesn’t much matter to most people in everyday life: people not appreciating how many mechanisms are involved in vision probably won’t affect their day or many other people’s days all that often. It’s effect on my life is certainly limited. However, when people begin discussing matters of morality, the effects of that instinct blindness can be a touch more important.

Blindness? Touch? I’ll see myself out…

The moral issue I would like to highlight today is one that I have hit upon many times in the past: gender. Specifically, the issue is that men and women are often found to differ with respect to particular outcomes to some degree: perhaps women, on the whole, tend to make less money than men; perhaps men tend to be sent to jail more often than women, and so on. Now it’s often also the case that people find such differences to be morally offensive. Well, sort of, anyway; more precisely, it’s not that the differences per se are morally offensive, but that the underlying causes of the differences are viewed to be nefarious in some way. It’s not that women make less money than men that is the problem, then, but rather that this fact is perceived to be due to sexism or discrimination against women that’s the problem.

I would like to emphasis the word “perceived” in that last sentence because claims of discrimination or sexism are often made without good supporting evidence, or their extent is, what one might consider, exaggerated to some degree. This isn’t to say that there is no such thing as discrimination or that sexism necessarily plays a minimal or no role in any given disparity, mind you; I don’t want to be misunderstood in that respect. The issue I’m discussing is that when people say things like “women make 70 cents for every dollar a man makes”, the implication being made, implicitly or explicitly, is that this 30 cent difference is due mostly or entirely to sexism and discrimination without consideration that any other factors might play some role in determining who makes how much money. Also, the implication is that such gaps should be reduced, of course. People aren’t just stating these gaps as if they were mere statements of facts; they’re calls to action.

Except that this clearly isn’t the case all the time, which brings us to the current paper (which looks like more of a conference presentation, but that’s besides the point). While not empirical in nature, the paper by Browne (2013) focuses on the following suggestion: gender differences that appear to favor men are far more often to be viewed as “gaps” requiring remediation, while gender differences that appear to favor women are viewed more as “facts” and of little or no moral or social concern. Browne (2013) runs through a few interesting examples of these disparities, among which are: the special focus on violence against women despite men being more likely to be a victim of almost any type of violent crime, women being less likely to be stopped or cited for traffic violations, women being sentenced to less time in jail if convicted of a crime, domestic abuse allegations of men being ignored at greater frequency than women’s, women earning more of the degrees than men in the US, and men making up a bit more than 9 out of every 10 workplace deaths. Despite the existence of these gender disparities, very little seems to ever be mentioned about them nor is much remedy for them sought; they seem to be viewed, more or less, as acceptable, or the unintended result of a system designed to benefit men overall.

“Can you reframe this workplace accident in the form of patriarchy?”

One of my favorite passages from the paper concerned research on one of the former issues: traffic stops. Though it’s lengthy, I wanted to recreate it here in its entirety because I think it demonstrates the focus of the paper rather aptly:

…[W]hen a Massachusetts study of racial and gender profiling found that, contrary to the authors’ expectations, women were substantially less likely to be stopped or cited than men, the authors did not then express concern that maybe there was gender profiling against men; instead, they emphasized the need for further information on “the traffic stop behavior of individual officers . . . to determine if some officers are stopping [a] larger number of female drivers compared to their similarly situated peers.” The fact that all officers, as a whole, were stopping a larger number of male drivers was simply not on the authors’ radar as a problem.

Such a passage suggests rather strongly that some research is conducted with a particular agenda in mind: the researchers seemed pretty sure that some group was being disadvantaged, and when they didn’t find the result they were looking for, they expressed interest in continuing to dig until they found the answer they wanted. The odds are good, I would say, that if the initial research turned up an identical gender “gap” disfavoring women (i.e. women are more likely to be stopped or cited for traffic offenses), it would be taken as evidence of a problem. But since this “gap” disfavored men, it was reported instead as more of a “fact”.

Research on our reasoning abilities has been reaching a similar conclusion for some time now: reasoning appears to function primarily to persuade other people of things, rather than to necessarily be accurate. Certain findings might be ignored or questions not asked if they don’t find the agenda of the researchers. Now it’s all well and good (and fun, too) to throw metaphorical rocks at the research or conclusions of other people and make accusations of particular agendas working against the empirical or theoretical soundness of their work. However, the interesting focus of this issue, to me, anyway, is not that people have biases, but rather why people have certain biases. Despite how many psychologists write on the topic, noticing (or labeling) a bias is not the same thing as explaining it. Something about gender – or some factor relating to it – seems to have a powerful, if perhaps under-appreciated or unrecognized, influence on our moral judgments. Why, then, might women’s welfare appear to be, in general, of more concern than men’s?

The answer to this question, I imagine, will likely turn out to be strategic in nature. Specifically, such a cognitive bias should only be expected to exist if it serves some other useful goal. The underlying logic here is that being wrong about reality can frequently carry costs, and these costs need to be offset by some compensating benefit in order for biases to persist and become common. So what might this other useful goal be? Well, I don’t think current accounts of our moral sentiments have much to offer us in that regard. The accounts of morality that suggest our moral psychology functions to increase group welfare or make people more altruistic/cooperative don’t seem to get us very far, as they don’t straightforwardly explain why one particular subgroup’s welfare (women) is more important than another’s (men). The dynamic coordination account – which posits that people take sides in moral disputes on the basis of observable actions to achieve coordination and reduce punishment costs (DeScioli & Kurzban, 2013) – doesn’t seem to get us very far either for two reasons: first, the model explicitly rejects the idea that moral sentiments should be tied to individual identities, so it’s not at all clear why one gender’s issues might be perceived differently and, second, because the observed part – the gender disparity – is not what people seem to be condemning; they are condemning the perceived cause of that disparity, and their perceptions may well be inaccurate on that front).

Because we know gender disparities can never be caused by choice; only sexism against women.

It seems more plausible to me that the selective attention and moral outrage that gets directed against particular gender “gaps” relates more to managing one’s association value to others. That is to say that supporting someone on a moral issue relates more to alliance politics than it does coordination or altruism. If, for instance, women happen to possess some resource (such as their reproductive capacity) that makes them more valuable socially (relative to non-women), then you might well find that people are, in general, more interested in catering to their issues. Even if one is not personally interesting in catering to those issues, however, if enough other people happen to be on the “women’s side” (provided such a term is meaningful, which I don’t think it is, but let’s use it anyway), siding against them can be a bad idea all the same: by doing so you might become a target of condemnation by proxy, even if you have personally done nothing particularly wrong, as you are preventing that group from achieving its goal. Now all of this speculation is founded on the idea that these “pro-women” biases actually exist, and I think that requires more empirical work to be demonstrated with greater certainty, but the anecdotes reviewed by Browne (2013) provide some good initial reasons to think such a phenomenon may well be real.

References: Browne, K. (2013). Mind which gap? The selective concern over sex disparities. Florida International Law Review, 8.

DeScioli, P. & Kurzban, R. (2013). A solution to the mysteries of morality. Psychological Bulletin, 139, 477-496.

What Does Online Dating Tell Us About Racial Views?

Preferences can be funny things, or at least our judgments of them. If I were to state that, “I have no interest in hiring a black person to do this job”, I would receive more than a little condemnation for that view. If I were to state instead that, “I have have no interesting in dating a black woman”, I would likely still receive some condemnation, but probably less than for the first statement. Finally, if I were to state that, “I have no interest in dating a man”, I would receive very little, if any, condemnation for it, even from those who advocate strongly for gay rights. As one of my colleagues recently posed the question, “Why is discrimination based on reproductive / sexual preferences OK, but other forms of discrimination are not?” The issue of discrimination is one I’ve discussed before, considering why discrimination on the basis of standardized test scores is deemed to be appropriate, whereas discrimination of the basis of obesity is often not. So let’s turn our attention towards discrimination in the sexual realm today.

“Free?! I’d have to be an idiot not to find the Asian of my dreams!”

A recent post by Jenny Davis over at the Pacific Standard suggests that “Online dating shows us the cold, hard facts about race in America“. In her article, Jenny discusses some data released from a Facebook-based dating app that figures out which people are interested in which other people on some sexual or romantic level. The data is labeled “unfortunate” in some respects, because there appear to be winners and losers, and those winners and losers seem to break down along racial lines. When it comes to mating, it seems that everyone doesn’t get to join hands and cross the finish line at the same time so that we all end up with equally-high self-esteem (I know; I was shocked too).  To give you a sense for the data (and so you don’t have to click back and forth between links), here’s the breakdown of the response rates for people who are interested.

As anyone can clearly see, there are favorites. When it comes to the highest positive response rate, most women, regardless of their race, appear to favor white men, whereas most men, again, regardless of their race, tend to favor Asian women. In terms of the lowest response rate, women appeared to shun black men, whereas men tended to shun black women. Ouch. Jenny, using what I can only assume is that same “high-powered sociological lens” I’ve encountered before, concludes that this clearly demonstrates that race matters, and serves to counter accusations that we are living in a color-blind, post-racial world. As Jenny puts it we “fetishize Asian women while devaluing blacks”. Now tone doesn’t come across well through text-based communications at time, but neither “fetishize” nor “devalue” sound as if they have a particularly positive connotation to me. It sounds as if she’s condemning other people for their sexual preferences in that respect.

There are many comments to make about this, but let’s start with this one: apparently, there’s something of a no-win situation being erected from the get go. When one group is preferred, it’s a “fetish”, whereas when they’re not preferred, they’re “devalued”. Well, sort of, anyway; if she were being consistent (and who is?) Jenny would also say that women “fetishize” white males. Strangely, she does not. One can only guess as to why she does not, because Jenny makes no apparent attempt to understand the data in question. By that, I mean that Jenny offers no potential alternative explanations through which we might understand the data. In fact, she doesn’t seem to offer any explanation whatsoever for these patterns of responses. If I had to, I would guess that her explanation, if simplified somewhat, would reduce to “racism did it”, but it’s hard to tell.

“But are they the Black singles of my dreams, like the Asians?”

I would like to try and pick up some of that explanatory slack. Despite initial appearances, it is possible that this data has very little, if anything, to do with race per se. Now I happen to think that race likely does matter to some extent when it comes to dating preferences, but the degree of that extent is anyone’s guess. To see why I would say this only requires that one understands a very basic statistical concept: correlation does not equal causation. This is something that I imagine Jenny understands, but it likely slipped her mind in the midst of trying to make a point. There are few examples to consider, but the first is by far the simplest. Most men, if you polled them, would overwhelming respond to women on dating websites, and not other men; women would likely do the reserve. This does not mean, however, that men (or women) “devalue” other men (or women). Similarly, just because people on these dating sites might respond to black people at the lowest rates, it does not mean they “devalue” black people more generally.

But maybe we do devalue certain racial groups, at least when it comes to dating them. This brings us to the second issue: mating decisions are often complex. There are dozens of potential variables that people assess when choosing a mate – such as how much money they have, how much they weigh, how tall they are, their age, their relatedness to us, etc – and the importance of these qualities also varies somewhat depending on the nature of the relationship (whether it is more short- or long-term, for instance). The important point here is that even if people are picking mates on the basis of these other characteristics alone and not race, we might still see racial differences in outcomes. Let’s say, for instance, that men tend to prefer women shorter than themselves as dating partners (the reasons for this preference or it’s actual existence need not necessarily concern us). If that were the case, provided there are any average differences in height among the races, we would still see different response rates to and from each racial group, even though no one was selecting on the basis of race.

Rather than just considering the direction the preferences in the data above, then, let’s consider some of the actual numbers: when it came to response rates, regardless of whether we were considering men or women, and regardless of whether we’re considering the highest or lowest response rates, black individuals seem to respond more often than any other group; sometimes around twice as often. This could be indicative of a number of different factors, though I won’t speculate as to which ones on the basis of the numbers alone. The only point is that those factors might show up in user’s profiles in some way. If other people pick up on those factors primarily, then race itself might not be the primary, or even a, factor driving these decisions. In fact, in terms of response rates, there was a consistent overall pattern: from lowest to highest, it tended to be Latinos, Whites, Asians, and Blacks, regardless of sex (with only a single exception). Whatever the reasons for this, I would guess that it shows up in other ways in the profiles of these senders and responders.

Strangely, I can’t find a picture of a white dating site. Odd…

As I said, I don’t think that race per se is entirely unrelated to mating choices. However, to determine the extent to which it uniquely predicts anything, you need to control for other relevant factors. Does obesity play a role in these decisions? Probably. Is obesity equally common across racial groups? Nope. How about income; does income matter? In some cases it sure seems to. Is income the same across racial groups? Nope. We would likely find the same for many, many other factors.

In addition to determining the extent of how much race matters, one might also wish to explain why race might matter. Simply noting that there appear to be some racial differences doesn’t tell us a whole lot; the same goes for correlations of match percentages and response rates over at OkCupid, which find a similar pattern with respect to race. In the instance of OkCupid, a match percentage of 10% between two people corresponds to about a 25% reply rate; a 90% match percentage gets you all the way up to… a 37% reply rate. Even at around 100% match, the response rate still only lingers at around 50%. There appears to be a lot more that goes into mating decisions than people typically appreciate or even recognize. For what it’s worth, I would rather work to understand those complexities than pat myself on the back for how bad I think racism is.

Begging Questions About Sexualization

There’s an old joke that goes something like this: If a man wants to make a woman happy, it’s really quite simple. All he has to do is be a chef, a carpenter, brave, a friend, a good listener, responsible, clean, warm, athletic, attractive, tender, strong, tolerant, understanding, stable, ambitious, and compassionate. Men should also not forget to compliment a women frequently, give her attention while expecting little in return, give her freedom to do what she wants without asking too many questions, and love to go shopping with her, or at least support the habit. So long as a man does/is all those things, if he manages to never forget birthdays, anniversaries, or other important dates, he should be easily able to make a woman happy. Women, on the other hand, can also make men happily with a few simple steps: show up naked and bring beer. (For the unabridged list, see here). While this joke, like many great jokes, contains an exaggeration, it also manages to capture a certain truth: the qualities that make a man attractive to a woman seem to be a bit more varied than the qualities that make a woman attractive to a man.

“Yeah; he’s alright, I guess. Could be a bit taller and richer…”

Even if men did value the same number of traits in women that women value in men, the two sexes do not necessarily value the same kinds of traits, or value them to the same degree (though there is, of course, frequently some amount of overlap). Given that men and women tend to value different qualities in one another, what should this tell us about the signals that each sex sends to appeal to the other? The likely answer is that men and women might end up behaving or altering their appearance in different ways when it comes to appealing to the opposite sex. As a highly-simplified example, men might tend to value looks and women might tend to value status. If a man is trying to appeal to women under such circumstance, it does him little good to signal his good looks, just as it does a woman no favors to try and signal her high status to men.

So when people start making claims about how one sex – typically women – are being “sexualized” to a much greater extent than the other, we should be very specific about what we mean by the term. A recent paper by Hatton & Trautner (2011) set forth to examine (a) how sexualized men and women tend to be in American culture and (b) whether that sexualization has seen a rise over time. The proxy measure they made use of for their analysis were about four decades worth of Rolling Stone covers, spanning from 1967 to 2009, as these covers contain pictures of various male and female cultural figures. The authors suggest that this research has value because of various other lines of research suggesting that these depictions might have negative effects on women’s body satisfaction, men’s negative attitudes about women, as well threatening to increase the amount of sexual harassment that women face. Somewhat surprisingly, in the laundry list of references attesting to these negative effects on women, there is no explicit mention of any possible negative effects on men. I find that interesting. Anyway…

As for the research itself, Hatton & Trautner (2011) examined approximately 1000 covers of Rolling Stone, of which 720 focused on men and 380 focused on women. The pictures were coded with respect to (a) the degree of nudity, from unrevealing to naked on a 6-point scale, (b) whether there was touching, from none to explicitly sexual on a 4-point scale, (c) pose, from standing upright to explicitly sexual on a 3-point scale, (d) mouth…pose (I guess), from not sexual to sexual on a 3-point scale, (e)  whether breasts/chest, genitals, or buttocks were exposed and/or the focal point of the image, all on 3-point scales, (f) whether the text on the cover line related to sex, (g) whether the shot focused on the head or the body, (h) whether the model was engaged in a sex act or not, and finally (i) whether there were hints of sexual role play suggested at. So, on the one hand, it seems like these pictures were analyzed thoroughly. On the other, however, consider this list of variables they were assessing and compare them to the initial joke. By my count, all of them appear to fall more on the end of “what makes men happy” rather than “what makes women happy”.

Which might cause a problem in translation from one sex to the other

Images were considered to be “hypersexualized” if they scored 10 or more points (out of the possible 23), but only regular “sexualized” if they scored from 5 to 9 points. In terms of sexualization, the authors found that it appeared to be increasing over time: in the ’60s, 11% of men and 44% of women were sexualized; by the ’00s these rose to 17% and 89% respectively. So Hatton & Trautner (2011) concluded that men were being sexualized less than women overall, which is reasonable given their criteria. However, those percentages captured both the “sexualized” and “hypersexualized” pictures. Examining the two groups separately, the authors found that around 1-3% of men on the covers were hypersexualized in any given decade, whereas the comparable range for women was 6% to 61%. Not only did women tend to be sexualized more often, they also tended to sexualized to a great degree. The authors go so far as to suggest that the only appropriate label for such depictions of women were as sex objects.

The major interpretative problem that is left unaddressed by Hatton & Trautner (2011) and their “high-powered sociological lens”, of course, is that they fail to consider whether the same kinds of displays make men and women equally sexually appealing. As the initial joke might suggest, men are unlikely to win many brownie points with a prospective date if they showed up naked with beer; they might win a place on some sex-offender list though, which falls short of the happy ending they would have liked. Indeed, many of the characteristics highlighted in the list of ways to make a woman happy – such as warmth, emotional stability, and listening skills – are not quite as easily captured by a picture, relative to physical appearance. To make matters even more challenging for the interpretation of the authors, there is the looming fact that men tend to be far more open to offers of casual sex in the first place. In other words, there might about as much value to signaling that a man is “ready for sex” as there is to signaling that a starving child is “ready for food”. It’s something that is liable to be assumed already.

To put this study in context, imagine I was to run a similar analysis to the authors, but started my study with the following rationale: “It’s well known that women tend to value the financial prospects of their sexual partners. Accordingly, we should be able to measure the degree of sexualization on Rolling Stone covers by assessing the net wealth of the people being photographed”.  All I would have to do is add in some moralizing about how depiction of rich men is bad for poorer men’s self-esteem and women’s preferences in relationships, and that paper would be a reasonable facsimile to the current one. If this analysis found that the depicted men tended to be wealthier than the depicted women, this would not necessarily indicate that the men, rather than the women, were being depicted as more attractive mates. This is due to the simple, aforementioned fact, that we should expect an interaction between signalers and receivers. It doesn’t pay for a signaler to send a signal that the intended receiver is all but oblivious to: rather, we should expect the signals to be tailored to the details of the receptive systems it is attempting to influence.

The sexualization of images like this might go otherwise unnoticed.

It seems that the assumptions made by the authors stacked the deck in favor of them finding what they thought they would. By defining sexualization in a particular way, they partially begged their way to their conclusion. If we instead defined sexualization in other ways that considered variables beyond how much or what kind of skin was showing, we’d likely come to different conclusions about the degree of sexualization. That’s not to say that we would find an equal degree of it between the sexes, mind you, but it would be a realization that there are many factors that can go into making someone sexually attractive which are not always able to be captured in a photo. We’ve seen complaints of sexualization like these leveled against the costumes that superheroes of various sexes tend to wear, and the same oversight is present in them as well. Unless the initial joke would work just as well if the sexes were reversed, these discussions will require more nuance concerning sexualization to be of much profitable use.

References: Hatton E. & Trautner, M. (2011). Equal opportunity objectification? The sexualization of men and women on the cover of Rolling Stone. Sexuality and Culture, 15, 256-278.