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.

Some Fuss Over Sperm Competition

Say what you will about Greg Cochran, but one thing is for certain: he’s certainly not one to present his opinions meekly. Recently, Greg posted his feelings about the relevance of sperm competition in humans, and he is of the mind that sperm competition is not terribly important for humans. For those of you not in the know, sperm competition refers to conditions under which a female has mated with more than one male at a time during which she might conceive. As the name might suggest, the sperm from different males can be thought of as “competing” to fertilize the egg in question (or eggs, depending on the species). The question of interest, for our present purposes, then, is whether or not such conditions (a) might have existed in ancestral human populations and (b) have been important enough to potentially drive male adaptations for solving the problem and winning the competition.

…And when they’re not, cheat to win.

Greg begins his post by making, what I feel, is an poor point, writing: “The non-paternity rate is an upper limit to the rate of sperm competition”. Unless I’m misunderstanding his meaning here, the rate of non-paternity (when a child is being unknowingly raised by a male that is not their father) would not be not the upper limit for sperm competition unless every single instance of sperm competition resulted in non-paternity. Presumably, there are many instances of sperm competition that the in-pair male would win, meaning that the non-paternity rate would be an underestimate of how much sperm competition there might be. The upper limit of sperm competition should, it seems, be the infidelity rate (how often men are being cheated on by their partners) or the number of times when report having sex with more than one male within the period of a few days. Depending on what numbers one wants to use there, the potential amount of sperm competition that might exist can go up dramatically.

On the other hand, however, not every instance of non-paternity results from sperm competition: sometimes a female might cheat on her partner while also not sleeping with him, allowing the rival’s sperm a competition-free environment. In such cases, non-paternity rates would overestimate the amount of sperm competition that exists. As to which set of issues are more common, I can’t say. There’s also the matter of how contraceptives and abortion might affect the issue, but I won’t consider them further. So while the non-paternity data is certainly informative in some sense, it’s far from what we might consider a complete picture when it comes to sperm competition. In any case, I’ll use Greg’s numbers of approximately 2% non-paternity on the whole across human populations and across time, just for the sake of argument.

On top of “how common is sperm competition”, then, another question we want to consider to get a full picture of the issue is, “how important is non-paternity?” As Greg notes, humans aren’t built for sperm competition in the way that, say, a chimpanzee male is. However, the corollary question he fails to consider concerns whether or not the consequences of non-paternity are identical between humans chimps. For a chimp, male paternal investment tends to reach the heights of not killing infants; for human males, investment might involve decades of protecting and provisioning. On that front, while pair-bonded human males might be far more assured of their paternity than your average chimp, the consequences of being mistaken that respect are also far larger for humans than chimps. In humans, a little bit of sperm competition goes a long way, so to speak. In such cases, a simple comparison between how much sperm competition exists will miss the bigger picture.

“Why’s everyone so upset? It’s just one guy that got shot…”

So we should expect the non-paternity rate to be low, in some absolute sense, as the consequences to non-paternity are so high; if non-paternity was too high, pair-bonding strategies would be unlikely to evolve in the first place, or persist once they had. Now, as Greg also mentions, there are likely adaptations humans possess to deal with the non-paternity issue that do not deal with sperm competition. As Greg so eloquently put it:

If, for example, your old lady knows that you will knock her block off if she strays, that deters all kinds of paternal uncertainty, not just those involving sperm competition. Your jealousy might also deter other guys from trying – adaptations for sperm competition don’t do deterrence.

This is certainly true enough: if a female partner knows that her straying will lead to physical aggression or the withdrawal of investment, she might be pressured into not having that affair; the same goes for rival males. However, aggression is not always the smartest strategy, as aggression carries costs. That rival male you seek to deter might well be bigger and stronger than you, and on top of being cheated on, you might also end up with an ass-kicking if you tried anything. Women may also have friends and family that prefer you didn’t hurt her, thank you very much.  So while sperm competition doesn’t serve a deterrent function, it also avoids costs associated with the aggressive deterrent function. Further, if deterrence fails, for whatever reason, sperm competition might also be able to serve as a secondary buffer against the non-paternity outcome.

This also raises the follow question: is non-paternity low, at least in part, because opportunities for sperm competition are rare or are they low because counter strategies by pair-bonded males are relatively effective? Perhaps the non-paternity rate might be substantially higher if in-pair males had no way of effectively mitigating the risks of sperm competition from rival males. After all, given that any particular sexual act is unlikely to result in conception, 2% non-paternity implies there were likely many more opportunities for non-paternity that were not realized. Now, to be clear, I’m not saying that we should expect human males to be absolutely chock-full of adaptations for sperm competition, nor am I suggesting that we show evidence of one particular strategy or another. I just want to point out some of the nuances of the argument for or against it that I think Greg gets wrong, or at least fails to discuss.

Adaptations for sperm competition might be more subtle than larger testicles, for instance. Perhaps the frequency of sex – or at least the frequency and intensity of sexual interest – correlates with infidelity cues; perhaps the number of sperm per ejaculate could be varied facultatively as a function of sperm competition risk. As Shackelford, Pound, & Goetz (2005) put it, humans might not show some hallmarks of adaptations designed for persistently-high levels of sperm competition (as such contexts were not persistent themselves, as they might be in chimps, for instance), but might show evidence of adaptations designed for contexts in which sperm competition risk is temporarily elevated. These adaptations might not be readily detected, but to write off their existence entirely on the basis of low-non-paternity rates itself would be premature.

“I wish my wife wasn’t such a premature climaxer…”

The question in my mind is not one of whether sperm competition matters or not for human populations, but rather to what degree it has. Dichotomizing such variables (“sperm competition was/was not important, depending on what precisely important means”) is unlikely to help us get a full picture of sperm competition in humans specifically or other species more generally. After all, chimps, humans, and gorillas were all descendent from a common ancestral species, and that species was unlikely to display the full range of mating behaviors evidence by all it’s future relatives. Some small initial degree of sperm competition must have been sufficient to get the ball rolling on later adaptations for such conditions. Now perhaps humans have been facing less sperm competition over time, and what we see are the degraded remnants of previous adaptations designed to deal with more of it. Then again, perhaps our species has gone the other way, or perhaps we have some adaptations designed for some modest levels of sperm competition. To be sure, some hypothetical adaptations for dealing with sperm competition will be wrong, and people’s estimates of how common it is or isn’t might be way off. I just wouldn’t close the door on the matter (or claim, as Greg does, that such adaptations “don’t exist”) because of that.

References: Shackelford, T., Pound, N., & Goetz, A. (2005). Psychological and physiological adaptations to sperm competition in humans. Review of General Psychology, 9, 228-248.

Dante’s Inferno

I’d like to begin with a quick apology for the tardiness of this latest update. It’s not that there’s anyone holding me to my usual weekly schedule except myself, but I am disappointed I haven’t gotten around to updating sooner. I had planned to take a week to myself to enjoy a new game (and enjoy it thoroughly I did, so mission accomplished there), but that week ended with my getting sick for another one, and I haven’t been able to concentrate on much as a result. With those excuses out of the way, I’m going to start off today like most middle/high-school students: by summarizing part of a book (or epic poem, really) that I haven’t read personally. Instead, I’ll be summarizing it – or at least part of it – by using the Wikipedia cliff notes. That story, as the title suggests, is Dante’s Inferno. What I really like about this Wikipedia page describing the story is that Inferno is kind enough to order the circles of hell for the reader with respect to increasing wickedness; the deeper one goes, the worse the sins needed to get there. The reason I like this neat and tidy ordering is that it gives us some insight as to the author’s moral sense.

I might not have read the book, but I did play the video game. Close enough.

As a quick rundown of the circles of hell, from least bad to worst, there’s: Limbo, Lust, Gluttony, Greed, Wrath, Heresy, Violence, Fraud, and finally Treachery. Now what’s particularly interesting here is that, according to Dante, it would seem to be worse to be a flatterer or a corrupt politician than a murderer. Misrepresenting your stances about people or politics is bad bad bad in Dante’s book. More interesting still is the inner-most circle: Treachery. Treachery seems to represent a particular kind of fraud: one in which the victim is expected to have some special relationship to the perpetrator. For instance, family members betraying each other seems to be worse than strangers doing similar harms. In general, kin are expected to behave more altruistically towards each other, owing in no small part to the fact that they share genes in common with one another. Helping one’s kin, in the evolutionary-sense of things, is quite literally like helping (part of) yourself. So if kin are expected to trade off their own welfare for family members at a higher rate than they would for strangers, but instead display the opposite tendency, this makes kin-directed immoral acts appear particularly heinous.

Now, of course, Dante’s take on things isn’t the only game in town. A paper which I have repeatedly discussed (DeScioli & Kurzban, 2013) has a different take on the issue of morality. That take is that morality serves, more or less, a coordination function for punishers: the goal is to get most people in agreement about who should be punished in order to avoid the fighting costs that are associated with disagreement in that realm. In order for this coordination function to work, however, the pair suggest that morality needs to function on the basis of acts; not the identity of the actors. As DeScioli & Kurzban (2013) put it:

“The dynamic coordination theory of morality holds that evolution favored individuals equipped with moral intuitions who choose sides in conflicts based, in part, on “morality” rather than relationship or status”

Identity shouldn’t come into play when it comes to moral condemnation, then; it is “[crucial that the signal] must not be tied to individual identity”. As Monty Python put it, “let’s not bicker and argue about who killed who“, and let’s not do that because killing should be equally as wrong no matter who does it and who ends up on the receiving end.

Now in fairness to DeScioli & Kuzrban (2013), they also hedge their theoretical bets, suggesting that identity also ought to matter when it comes to picking sides in disputes. However, it seems that, according to the dynamic coordination model, anyway, when people do take sides on the basis loyalty to their friends or family, they should be motivated by systems that do not deal with morality. This suggestion seems to be at odds with Dante’s less-formalized interpretation of the importance of identity in the realm of morality, who instead would appear to hypothesize, at least implicitly, that the identity of the actors ought to matter a great deal. So let’s take a look at some research bearing on the matter.

And let’s do so quickly, before I get back to being addicted to this game.

The first piece of research comes to us from Lieberman & Linke (2007) who were examining whether the identity of an individual (either a foreigner, schoolmate, or family member) mattered when it came to the wrongness of an act and amount of punishment deemed to be appropriate for it (in this case, stealing $1500). When the individual in question was the perpetrator, participants (N = 268) suggested that the foreigner deserved more punishment than the schoolmate, and that the schoolmate deserved more punishment than the family member. Family members were also perceived to be more remorseful about their act, relative to schoolmates, relative to strangers. On the other hand, people’s rating of the immorality of the act did not vary as a function of the actor’s identity; no matter who one was, the act was rated as just as morally wrong (though ratings in all cases were close to ceiling levels here).

The next experiment (N = 288) examined essentially the same question, but this time the individual in question was the victim of the offense, rather than the perpetrator. When the offense was committed against a family member, people tended to be more punitive towards the perpetrator than when it was committed against a schoolmate or foreigner. Again, however, moral judgments remained uniformly at ceiling levels in all cases. In a final experiment (N = 78) participants were asked about how much they would be willing to personally invest in order to track down the perpetrator of various deeds. As before, people reported being willing to take more days off from work without pay to try and find the thief when a family member had been robbed (M = 12.85 days), relative to a schoolmate (M = 2.24) or a foreigner (M = 2.10). Now whether or not people would actually do these things (I don’t recall many people taking time off work to play Batman and help strangers track down thieves), people are at least expressing sentiments indicating that they think people should be punished to a greater degree for victimizing their kin, and that their kin deserve less punishment.

The results could be taken to favor either account – Dante’s or DeScioli & Kurzban’s – I feel. On the one hand, rating of morality appeared stubbornly impartial: the act was rated as being just as morally wrong, no matter the identity of the perpetrator or victim. This might suggest people were coordinating around the behavior, and not the identity of the actors. However, people were also not coordinating their behavior in the sense that what they actually wanted to see done after they had decided the act was morally wrong varied on the basis of identity. To express this tension in a different context, we might consider the following: imagine that most people agree with the statement, “freedom is a good thing”; good for America. However, that certainly does not mean that most people would be in agreement when it came to what precisely that sentence is supposed to mean: that is, what limits are to be put in place, and how those limits should be enacted?

Just exercising his freedom to pepper-spray protesters.

That said, the paper by Lieberman & Linke (2007) doesn’t exactly get at what Dante was proposing: Dante didn’t, as far as I know, anyway, say that lust was any better or worse when a family member does it. After all, everyone is someone’s family member, or friend, or foreigner. Instead, what Dante appeared to be proposing is that the relationship of the perpetrator to the victim is the crucial variable. As I’ve discussed previously, some initial research has tentatively borne out Dante’s hypothesis: while acts are rated as morally worse than omissions between strangers, this difference is reduced when the interaction occurs between friends, and the act is rated as more morally wrong overall. A more formal test of these competing hypotheses appears to await data. I’ll be sure to get right on that personally; just as soon as I’m done being addicted to this game in the next three or four years.

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

Lieberman, D. & Linke, L. (2007). The effect of social category on third party punishment. Evolutionary Psychology, 5, 289-305.

When Are Equivalent Acts Not Equal?

There’s been an ongoing debating in the philosophical literature on morality for some time. That debate focuses on whether the morality of an act should be determined on the basis of either (a) the act’s outcome, in terms of its net effects on people’s welfare, or (b) whether the morality of an act is determined by…something else; intuitions, feelings, or what have you (i.e. “Incest is just wrong, even if nothing but good were to come of it”). These stances can be called the consequentialist and nonconsequentialist stances, respectively, and it’s at topic I’ve touched upon before. When I touched on the issue, I had this to say:

There are more ways of being consequentialist than with respect to the total amount of welfare increase. It would be beneficial to turn our eye towards considering strategic welfare consequences that likely to accrue to actors, second parties, and third parties as a result of these behaviors.

In other words, moral judgments might focus not only on the acts per se (the nonconsequentalist aspects) or their net welfare outcomes (the consequences), but also on the distribution of those consequences. Well, I’m happy to report that some very new, very cool research speaks to that issue and appears to confirms my intuition. I happen to know the authors of this paper personally and let me tell you this: the only thing about the authors that are more noteworthy than their good looks and charm is how humble one of them happens to be.

Guess which of us is the humble one?

The research (Marczyk & Marks, in press) was examining responses to the classic trolley dilemma and a variant of it. For those not well-versed in the trolley dilemma, here’s the setup: there’s an out-of-control train heading towards five hikers who cannot get out of the way in time. If the train continues on it’s part, then all five wills surely die. However, there’s a lever which can be pulled to redirect the train onto a side track where a single hiker is stuck. If the lever is pulled, the five will live, but the one will die (pictured here). Typically, when asked whether it would be acceptable for someone to pull the switch, the majority of people will say that it is. However, in past research examining the issue, the person pulling the switch has been a third party; that is, the puller was not directly involved in the situation, and didn’t stand to personally benefit or suffer because of the decision. But what would happen if the person pulling the switch was one of the hikers on one of the tracks; either on the side track (self-sacrifice) or the main track (self-saving)? Would it make a difference in terms of people’s moral judgments?

Well, the nonconsequentist account would say, “no; it shouldn’t matter”, because the behavior itself (redirecting a train onto a side track where it will kill one) remains constant; the welfare-maximizing consequentialist account would also say, “no; it shouldn’t matter”, because the welfare calculations haven’t changed (five live; one dies). However, this is not what we observe. When asked about how immoral it was for the puller to redirect the train, ratings were lowest in the self-sacrifice condition (M = 1.40/1.16 on a 1 to 5 scale in international and US samples, respectively), in the middle for the standard third-party context (M = 2.02/1.95), and highest in the self-saving condition (M = 2.52/2.10). In terms of whether or not it was morally acceptable to redirect the train, similar judgments cropped up: the percentage of US participants who said it was acceptable dropped as self-interested reasons began to enter into the question (the international sample wasn’t asked this question). In the self-sacrifice condition, these judgments of acceptability were highest (98%), followed by the third-party condition (84%), with the self-saving condition being the lowest (77%).

Participants also viewed the intentions of the pullers to be different, contingent on their location in this dilemma: specifically, the more one could benefit him or herself by pulling, the more people assumed that was the motivation for doing so (as compared with the puller’s motivations to help others: the more they could help themself, the less they were viewed as intending to help others). Now that might seem unsurprising: “of course people should be motivated to help themselves”, you might say. However, nothing in the dilemma itself spoke directly to the puller’s intentions. For instance, we could consider the case where a puller just happens to be saving their own life by redirecting the train away from others. From that act alone, we learn nothing about whether or not they would sacrifice their own life to save the lives of others. That is, one’s position in the self-beneficial context might simply be incidental; their primary motivation might have been to save the largest number of lives, and that just so happens to mean saving their own in the process. However, this was not the conclusion people seemed to be drawing.

*Side effects of saving yourself include increased moral condemnation.

Next, we examined a variant of the trolley dilemma that contained three tracks: again, there were five people on the main track and one person on each side track. As before, we varied who was pulling the switch: either the hiker on the main track (self-saving) or the hiker on the side track. However, we now varied what the options of the hiker on the side track were: specifically, he could direct the train away from the five on the main track, but either send the train towards or away from himself (the self-sacrifice and other-killing conditions, respectively). The intentions of the hiker on the side track, now, should have been disambiguated to some degree: if he intended to save the lives of others with no regard for his own, he would send the train towards himself; if he intended to save the lives of the hikers on the main track while not harming himself, he would send the train towards another individual. The intentions of the hiker on the main track, by contrast, should be just as ambiguous as before; we shouldn’t know whether that hiker would or would not sacrifice himself, given the chance.

What is particularly interesting about the results from this experiment is how closely the ratings of the self-saving and other-killing actors matched up. Whether in terms of how immoral it was to direct the train, whether the puller should be punished, how much they should be punished, or how much they intended to help themselves and others, ratings were similar across the board in both US and international samples. Even more curious is that the self-saving puller – the one whose intentions should be the most ambiguous – was typically rated as behaving more immorally and self-interestedly – not less – though this difference wasn’t often significant. Being in a position to benefit yourself from acting in this context seems to do people no favors in terms of escaping moral condemnation, even if alternative courses of actions aren’t available and the act is morally acceptable otherwise.

One final very interesting result of this experiment concerned the responses participants gave to the open-ended questions, “How many people [died/lived] because the lever was pulled?” On a factual level, these answers should be “1″ and “5″ respectively. However, our participants had a somewhat different sense of things. In the self-saving condition, 35% of the international sample and 12% of the US sample suggest that only 4 people were saved (in the other-killing condition, these percentages were 1% and 9%, and in the self-sacrifice condition they were 1.9% and 0%, respectively). Other people said 6 lives had been saved: 23% and 50% in the self-sacrifice condition, 1.7% and 36% in the self-saving condition, and 13% and 31% in the (international and US respectively). Finally, a minority of participants suggested that 0 people died because the train was redirected (13% and 11%), and these responses were almost exclusively found in the self-sacrifice conditions. These results suggest that our participants were treating the welfare of the puller in a distinct manner from the welfare of others in the dilemma. The consequences of acting, it would seem, were not judged to be equivalent across scenarios, even though the same number of people actually lived and died in each.

“Thanks to the guy who was hit by the train, no one had to die!”

In sum, the experiments seemed to demonstrate that these questions of morality are not to be limited to considerations of just actions and net consequences: to whom those consequences accrue seems to matter as well. Phrased more simply, in terms of moral judgments, the identity of actors seems to matter: my benefiting myself at someone else’s expense seems to have much different moral feel than someone else benefiting me by doing exactly the same thing. Additionally, the inferences we draw about why people did what they did – what their intentions were – appear to be strongly affected by whether that person is perceived to have benefited as a result of their actions. Importantly, this appears to be true regardless of whether that person even had any alternative courses of action available to them. That latter finding is particularly noteworthy, as it might imply that moral judgments are, at least occasionally, driving judgments of intentions, rather than the typically-assumed reverse (that intentions determine moral judgments). Now if only there was a humble and certainly not self-promoting psychologist who would propose some theory for figuring out how and why the identity of actors and victims tends to matter…

References: Marczyk, J. & Marks, M. (in press). Does it matter who pulls the switch? Perceptions of intentions in the trolley dilemma. Evolution and Human Behavior.

Who’s Blaming The Victim?

One phrase – or some variant of it – that seems to crop up in discussions of sexual assault more frequently than almost all others is “blaming the victim” (though I imagine “rape culture” and “patriarchy” are probably in the running for most commonly-used term as well). Coined in the early 1970s, the phrase has been nothing but gaining in popularity if Google’s N-gram viewer is any indication. The way I’ve seen the term used, “blaming the victim” appeared to amount to any suggestion that sexual assault might be reduced through any behavioral modifications on the part of victims of the act; in other words, any suggestion that people bear some responsibility for ensuring their own safety. Now, of course, sexual assault victims are no more at moral fault for the crime they suffer than robbery victims are at fault for being robbed (which is to say not morally at fault at all). The responsibility for the criminal act lies at the feet of the criminal. Nevertheless, one might also responsibly suggest that precautions might be taken to minimize the frequency of such acts, in much the same way that criminals are at fault for stealing bikes, but one might suggest that people lock their bikes up so as to make theft more difficult.

“Good idea, but your execution of it leaves something to be desired.”

The recent Slutwalks were one of the more notable outcomes of such a suggestion: specifically, the suggestion that women might be able to minimize their risks of being sexually assaulted by dressing less provocatively. While I have no data on hand bearing on the plausibility of such a suggestion, I happen to have some other interesting research on the topic of victim blaming in the case of sexual assault. What’s unique about the current study by Perilloux et al (2014) is the examination of how different parties assign responsibility for a sexual assault: more precisely, how victims of a completed or attempted sexual assault assign responsibility, relative to third parties who were not assaulted themselves, but know a friend that was. The questions of interest here were (a) whether these three groups differ in terms of how much responsibility they assign to various parties, and (b) whether these groups also perceive the motivations of the attacker differently as well.

The sample included 49 women who self-reported experiencing a completed sexual assault after puberty, 91 women who reported an attempted sexual assault, and 152 women who reported knowing someone who was the victim of an assault. The participants were asked to assign blame (totaling 100%) for the assault to six potential sources: the perpetrator, the victim, the situation, the victim’s family, friends, or other categories. They were then asked to response in an open-ended fashion as to why they had assigned blame the way they did. The participants were also asked what they thought the perpetrators hoped to gain from the assault.

The results found some interesting disconnects between the perceptions of these groups. For the most part, the three groups – completed, attempted, and third parties – were in agreement over how much blame the situation, friends, family members, and other factors shared for the assault (approximately 7%, 2%, 1%, and 1%, respectively; so about 10% of the overall blame). Where these groups differed primarily was with respect to how much blame the victim of the assault and perpetrator share. Those women who were the victims of a completed or attempted assault suggested the perpetrator bore about 70% of the blame while they – the victim – were about 19% responsible. Third parties – those women who were not assaulted but knew someone who said they were – reported a different pattern: the third parties suggested the victim (their friends) was only about 9% responsible, while the perpetrator was 82% responsible. In other words, the victims themselves seemed to be doing about twice as much of the victim blaming than their friends were.

Or remember to always make accusations with all five fingers. Problem solved.

That wasn’t the only avenue along which these perceptions diverged, though: the three groups also differed in terms of how they perceived the attacker’s motivation (i.e. why the assaulter did what they did). In the completed assault group, 65% of women nominated “sex” as the primary motivation for the assault, while 22% suggested power was the motivation. These percentages were similar to the attempted group (71% and 18%, respectively). However, the third party women saw things rather differently: only 48% suggested sex was the motivation for the action, while 27% suggested power was driving the act. So the friends of the assault victims appeared to feel the assault was less about sex, relative to the women who were actually assaulted.

Finally, the analysis turned to only the perceptions of the completed and attempted groups. Perilloux et al (2014) examined the most common reasons listed for self-blame: (1) putting oneself in a bad situation, (2) being intoxicated, (3) not resisting enough, (4) sending mixed messages, and (5) being too trusting. Perhaps unsurprisingly, completed assault victims were more likely to list “not resisting enough” (25%) relative to attempted victims (11%), owing to the probability that resistance likely serves as a deterrent, and how much resistance is “enough” is assessed by whether or not the assault was stopped, or when it was. More completed victims (29%) also listed their own intoxication as a reason for their blame, relative to attempted victims (19%), and the completed group also reported more intoxication at the time of the assault. Again, this is might well be related to the resistance factor: intoxicated people could be less capable or willing to resist.

So, the good news from this research should be that, by in large, people seem to overwhelming place the blame for sexual assault on the perpetrator rather than the victim. The blame given to the perpetrator tended to be three- to nine-times that of the blame assigned to the victim. This appears to be true regardless of whether it’s the victim or the victim’s friend. The most interesting finding seems to be the disconnect between the responses of third parties and victims, however: most of the victim blame came from the victims themselves, and these victims tended to see the motivation for sex on the part of perpetrator as playing a more primary role than third parties did. There is, of course, the possibility that other, more socially-distant, third parties would assign more blame to the victims, relative to current groups, but that much remains to be seen. In any case, the question to consider is why these perceptions differ.

*Warning: point of view might not match well to reality.

One possibility is that the victims might have greater insight to what factors increased their risks for the assault, relative to third parties, owing to the fact that they were direct witnesses to the event. This certainly seems like a reasonable suggestion, and should give pause to those who claim that rape is primarily and act of violence or domination, rather than sex. Another, not mutually-exclusive suggestion, that I would advance would be to consider what signals these perceptions of victim blame might be sending. Since there’s no objective truth to the question, “how much blame does party X deserve”, these perceptions are likely to be reflecting something else.

Here are two possible alternatives as to what that something else might be: the first is that “who deserves how much blame?” might be interpreted as “whose side would you take in a dispute between the victim and the perpetrator?” In placing very little blame on the victim, third parties could be signaling a strong willingness to take their friend’s side on the matter. Another (also non-mutually exclusive) potential is that the question about who deserves blame might be interpreted as, “how much did the behavior of this individual increase their probability of being assaulted?” In this case, victims, through their self-blame, might be signaling that they recognize some potential for minimizing their future risk of being assaulted. This recognition could, in turn, make the victim look like a better social ally. Friends who consistently expose themselves to costly risks are, all else being equal, more costly to consistently support and side with than friends who suffer fewer costs. Accordingly, a friend who suggests they will behave more cautiously next time might appear to be at a lower risk for suffering costs, and a better social investment.

If third parties and victims interpret the notion of “blame” somewhat differently, then, this would lead to the following prediction: when rating one’s own blame for sexual assault, victims should rate their own blame higher, relative to third parties, as they did in the current study. However, there’s another prediction we could make: when rape victims are rating another victim’s blame for their sexual assault, the former group should not differ from non-victimized third parties. That is, victims of sexual assault should not both blame themselves and other victims equally; how much blame they assign to themselves or others should vary strategically.

References: Perilloux, C., Duntley, J., & Buss, D. (2014). Blame attribution in sexual victimization. Personality and Individual Differences, 63, 81-86.

Imagine Psychology Without People

In 1971, John Lennon released the now-iconic song “Imagine“. In the song, Lennon invites us to imagine a world without religion, countries, or personal possessions where everyone coexists in peace with one another. Now, of course, this is not the world in which we exist. In fact, Lennon apparently preferred to keep this kind of world in the realm of imagination himself, using his substantial personal wealth to live a life well-beyond his needs; a fact which Elton John once poked fun at, rewriting to lyrics to imagine to begin: “Imagine six apartments; it isn’t hard to do. One’s full of fur coats; the other’s full of shoes”. While Lennon’s song might appear to have an uplifting message (at least superficially; I doubt many of us would really want to live in that kind of world if given the opportunity), the message of the song does not invite us to understand the world as it is: we are asked to imagine another world; not to figure out why our world bears little resemblance to that one.

My imaginations may differ a bit from John’s, but to each their own.

Having recently returned from the SPSP conference (Society of Personality and Social Psychology), I would like to offer my personal reflections about the general state of psychological research from my brief overview of what I saw at the conference. In the sake of full disclosure, I did not attend many of the talks and I only casually browsed over most of the posters that I saw. The reason for this state of affairs, however, is what I would like to focus on today. After all, it’s not that I’m a habitual talk-avoider: at last year’s HBES conference (Human Behavior and Evolution Society), I found myself attending talks around the clock; in fact, I was actually disappointed that I didn’t get to attend more of them (owing in part to the fact that pools tend to conceal how much you’ve been drinking). So what accounted for the differences in my academic attendance at these two conferences? There are two particular factors I would like to draw attention to, which I think paint a good picture my general impressions of the field of psychology.

The first of these factors was the organization of the two conferences. At HBES, the talks were organized, more or less, by topics: one room had talks on morality, another on life history, the next on cooperation, and so on. At SPSP, the talks were organized, as far as I could tell, anyway, with no particular theme. The talks at SPSP seemed to be organized around whatever people putting various symposiums together wanted to talk about, and that topic tended to be, at least from what I saw, rather narrow in its focus. This brings me to the first big difference between the two conferences, then: the degree of consilience each evidenced. At HBES, almost all the speakers and researchers seemed to share a broader, common theoretical foundation: evolutionary theory. This common understanding was then applied to different sub-fields, but managed to connect all of them into some larger whole. The talks on cooperation played by the same rules, so to speak, as the talks on aggression. By contrast, the psychologists at SPSP did not seem to be working under any common framework. The result of this lack of common grounding is that most of these talks were islands unto themselves, and attending one of them probably wouldn’t tell you much about any others. That is to say that a talk at SPSP might give you a piece of evidence concerning a particular topic, but it wouldn’t help you understand how to think about psychology (or even that topic) more generally. The talks on self-affirmation probably wouldn’t tell you anything about the talks on self-regulation, which in turn bear little resemblance to talks on sexism.

The second big issue is related to the first, and where our tie in to John Lennon’s song arises. I want you to imagine a world in which psychology was not, by in large, the study of human psychology and behavior in particular, but rather the study of psychology among life in general. In this world we’re imagining, humans, as a species, don’t exist as far as psychological research is concerned.  Admittedly, such a suggestion might not lend itself as well to song as Lennon’s “Imagine”, but unlike Lennon’s song, this imagination actually leads us to a potentially useful insight. In this new world – psychology without people – I only anticipate that one of these two conferences would actually exist: HBES. The theoretical framework of the researchers at HBES can help us understand things like cooperation, the importance of kinship, signaling, and aggression regardless of what species we happen to be talking about. Again, there’s consilience when using evolutionary theory to study psychology. But what about the SPSP conference? If we weren’t talking about humans, would anyone seriously try to use concepts like the “glass ceiling”, “self-affirmation”, “stereotypes”, or “sexism” to explain the behavior of any non-human organisms? Perhaps; I’ve just never seen it happen.

“Methods: We exposed birds to a stereotype threat condition…”

Now, sure; plenty of you might be thinking something along the lines of, “but humans are special and unique; we don’t play by the same rules that all other life on this planet does. Besides, what can the behavior of mosquitoes, or the testicle size of apes tell us about human psychology anyway?” Such a sentiment appears to be fairly common. What’s interesting to note about that thought, however, would not only be that it seem to confirm that psychology suffers from a lack of consilience, but, more importantly, it would be markedly mistaken. Yes; humans are a unique species, but then so is every other species on the planet. It doesn’t follow from our uniqueness that we’re not still playing the same game, so to speak, and being governed by the same rules. For instance, all species, unique as they are, are still subject to gravitational forces. By understanding gravity we can understand the behavior of many different falling objects; we don’t need separate fields of inquiry as to how one set of objects falls uniquely from the others. Insisting that humans are special in this regard would be a bit like an ornithologist insisting that the laws of gravity don’t apply to most bird species because they don’t fall like rocks tend to. Similarly, all life plays by the rules of evolution. By understanding a few key evolutionary principles, we can explain a remarkable amount of the variance in the way organisms behave without needing disparate fields for each species (or, in the case of psychology, disparate fields for every topic).

Let’s continue to imagine a bit more: if psychology had to go forward without studying people, how often do you think would find people advocating suggestions like this:

If our university community opposes racism, sexism, and heterosexism, why should we put up with research that counters our goals simply in the name of “academic freedom”?…When an academic community observes research promoting or justifying oppression, it should ensure that this research does not continue.

Maybe in our imaginary world of psychological research without people there would be some who seriously suggested that we should not put up with certain lines of research. Maybe research on, say, the psychology of mating in rabbits should not be tolerated, not because it’s inaccurate, mind you, but rather because the results of it might be opposed to the predetermined conclusions of anti-rabbit-heterosexism-oppression groups. Perhaps research on how malaria seems to affect the behavior of mosquitoes shouldn’t be tolerated because it might be used to oppress mosquitoes with seemingly “deviant” or extreme preferences for human blood. Perhaps these criticisms might come up, but I don’t imagine such opposition would be terribly common when the topic wasn’t humans.

“Methods: We threatened the elephant seal’s masculinity…”

So why didn’t I attend as many talks at SPSP as I did at HBES? First, there was the lack of consilience: without the use or consideration of evolutionary theory explicitly, a lot of the abstracts for research at SPSP sounded as if they would represent more of an intellectual spinning of wheels rather than a forwarding of our knowledge. This perception, I would add, doesn’t appear to be unique to me; certain psychological concepts seem to have a nasty habit of decaying in popularity over time. I would chalk that up to their lack of being anchored to or drawn from some underlying theoretical concept, but I don’t have the data on hand to back that up empirically at the moment. The second reason I didn’t attend as many talks at SPSP was because some of them left me with the distinct sense that the research was being conducted with some social or political goal in mind. While that’s not to say it necessarily disqualifies the research from being valuable, it does immediately make me skeptical (for instance, if you’re researching “stereotypes”, you might want to test their accuracy before you write them off as a sign of bias. This was not done at the talks I saw).

Now all of this is not simply said in the service of being a contrarian (fun as that can be) nor am I saying that every piece of research to come out of an evolutionary paradigm is good; I have attended many low- to mid-quality talks and posters at the evolutionary conferences I’ve been to. Rather, I say all this because I think there’s a lot of potential for psychological research in general to improve, and the improvement itself wouldn’t be terribly burdensome to achieve. The tools are already at our disposal. If we can collectively manage to stop thinking of human behavior as something requiring a special set of explanations and start seeing it within a larger evolutionary perspective, a substantial amount of the battle will already be won. It just takes a little imagination.

Does Grief Help Recalibrate Behavior?

Here’s a story which might sound familiar to all of you: one day, a young child is wandering around in the kitchen while his parents are cooking. This child, having never encountered a hot stove before, reaches up and brushes his hand against the hot metal. Naturally, the child experiences a physical pain and withdraws his hand. In order to recalibrate his behavior so as to not avoid future harms, then, the child spends the next week unable to get out of bed – owing to a persistent low-energy – and repeatedly thinks about touching the hot stove and how sad it made him feel. For the next year, the child returns to the spot where he burned his hand, leaving flowers on the spot, and cries for a time in remembrance. OK; so maybe that story doesn’t sound familiar at all. In fact, the story seems absurd on the face of it: why would the child go through all that grief in order to recalibrate their stove-touching behavior when they could, it seems, simply avoid touching the hot stove again? What good would all that additional costly grief and depression do? Excellent question.

Unfortunately, chain emails do not offer learning trials for recalibration.

In the case of the hot stove, we could conclude that grief would likely not add a whole lot to the child’s ability to recalibrate their behavior away from stove-touching. It doesn’t seem like a very efficient way of doing so, and the fit between the design features of grief and recalibration seem more than a bit mismatched. I bring these questions up in response to a suggestion I recently came across by Tooby & Cosmides, with whom I generally find myself in agreement with (it’s not a new suggestion; I just happened to come across it now). The pair, in discussing emotions, have this to say about grief:

Paradoxically, grief provoked by death may be a byproduct of mechanisms designed to take imagined situations as input: it may be intense so that, if triggered by imagination in advance, it is properly deterrent. Alternatively-or additionally-grief may be intense in order to recalibrate weightings in the decision rules that governed choices prior to the death. If your child died because you made an incorrect choice (and given the absence of a controlled study with alternative realities, a bad outcome always raises the probability that you made an incorrect choice), then experiencing grief will recalibrate you for subsequent choices. Death may involve guilt, grief, and depression because of the problem of recalibration of weights on courses of action. One may be haunted by guilt, meaning that courses of action retrospectively judged to be erroneous may be replayed in imagination over and over again, until the reweighting is accomplished.

So Tooby and Cosmides posit two possible functions for grief here: (1) there isn’t a function per se; it’s just a byproduct of a mechanism designed to use imagined stimuli to guide future behavior, and (2) grief might help recalibrate behavior so as to avoid outcomes that previously have carried negative fitness consequences. I want to focus on the second possibility because, as I initially hinted at, I’m having a difficult time seeing the logic in it.

One issue I seem to be having concerns the suggestion that people might cognitively replay traumatic or grief-inducing events over and over in order to better learn from them. Much like the explanation often on offer for depression, then, grief might function to help people make better decisions in the future. That seems to be the suggestion Tooby & Cosmides are getting at, anyway. As I’ve written before, I don’t think this explanation in plausible on the face of it. At least in terms of depression, there’s very little evidence that depression actually helps people make better decisions. Even if it did, however, it would raise the question as to why people ever don’t make use of this strategy. Presumably, if people could learn better by replaying events over and over, one might wonder why we ever don’t do that; why would we ever perform worse, when we could be performing better?  In order to avoid making what I nicknamed the Dire Straits fallacy ( from their lyric “money for nothing and the chicks for free“), the answer to that question would inevitably involve referencing some costs to replaying events over and over again. If there were no such costs to replay, and replay led to better outcomes, replay should be universal, which it isn’t; at least not to nearly the same degree. Accordingly, any explanation for understanding why people use grief as a mechanism for improved learning outcomes would need to make some reference as to why grief-struck individuals are more able to suffer those costs for the benefits continuous replay provides. Perhaps such an explanation exists, but it’s not present here.

One might also wonder what replaying some tragic event over and over would help one learn from it. That is, does the replaying the event actually help one extract additional useful information from the memory? As we can see from the initial example, rumination is often not required to quickly and efficiently learn connections between behaviors and outcomes. To use the Tooby & Cosmides example, if your child died because you made an incorrect choice, why would ruminating for weeks or longer help you avoid making that choice again? The answer to that question should also explain why rumination would not be required for effective learning in the case of touching the hot stove.

It should only be a few more weeks of this until she figures out that babies need food.

One might also suggest that once the useful behavioral-recalibration-related information has been extracted from the situation, replaying the grief-inducing event would seem to be wasted time, so the grief should stop. Tooby & Cosmides make this suggestion, writing:

After the 6-18 month period, the unbidden images suddenly stop, in a way that is sometimes described as “like a fever breaking”: this would be the point at which the calibration is either done or there is no more to be learned from the experience

The issue I see with that idea, however, is that unless one is positing it can take weeks, months, or even years to extract the useful information from the event, then it seems unlikely that much of that replay involves helping people learn and extract information. Importantly, to the extent that decisions like these (i.e. “what were you doing that led to your child’s death that you shouldn’t do again”) were historically recurrent and posed adaptive problems, we should expect evolved cognitive decision making modules to learn from them fast and efficiently. A mechanism that takes weeks, months, or even years to learn from an event by playing it over and over again should be at a massive disadvantage, relative to a mechanism that can make those same learning gains in seconds or minutes. A child that needed months to learn to not touch a hot stove might be at a risk of touching the stove again; if the child immediately learned to not do so, there’s little need to go over grieving about it for months following the initial encounter. Slow learning is, on the whole, a bad thing which carries fitness costs; not a benefit. Unless there’s something special about grief-related learning that requires it takes so long – some particularly computationally-demanding problem – then the length of grief seems like a peculiar design feature for recalibrating one’s own behavior.

This, of course, all presumes that the grief-recalibration learning mechanisms know how to recalibrate behavior in the first place. If your child died because of a decision you made, there are likely very many decisions that you made which might or might not have contributed to that outcome. Accordingly, there are very many ways in which you might potentially recalibrate your behavior to avoid such a future outcome again, very few of which will actually be of any use. So your grief mechanism should need to know which decisions to focus on at a minimum. Further still, the mechanism would need to know if recalibration was even possible in the first place. In the case of a spouse dying from something related to old age or a child dying from an illness or accident, all the grieving in the world wouldn’t necessarily be able to effect any useful change the next time around. So we might predict that people should only tend to grieve selectively: when doing so might help avoid such outcomes in the future. This means people shouldn’t tend to grieve when they’re older (since they have less time to potentially change anything) or about negative outcomes beyond their control (since no recalibration would help). As far as I know (which, admittedly, isn’t terribly far in this domain) this isn’t that case. Perhaps an astute reader could direct me to research where predictions like these have been tested.

Finally, humans are far from the only species which might need to recalibrate their behavior. Now it’s difficult to say precisely as to what other species feel, since you can’t just ask them, but do other species feel grief the same way humans do? The grief-as-recalibration model might predict that they should. Now, again, the depth of my knowledge on grief is minimal, so I’m forced to ask these questions out of genuine curiosity: do other species evidence grief-related behaviors? If so, in what contexts are these behaviors common, and why might those contexts be expected to require more behavioral recalibration than non-grief-inducing situations? If animals do not show any evidence of grief-related behaviors, why not? These are all matters which would need to be sorted out. To avoid the risk of being critical without offering any alternative insight, I would propose an alternative function for grief similar to what Ed Hagen proposed for depression: grief functions to credibly signal one’s social need.

“Aww. Looks like someone needs a hug”

Events that induce grief – like the loss of close social others or other major fitness costs – might tend to leave the griever in a weakened social position. The loss of mates, allies, or access to resources poses major problems to species like us. In order to entice investment from others to help remedy these problems, however, you need to convince those others that you actually do have a legitimate need. If your need is not legitimate, then investment in your might be less liable to payoff. The costly extended periods of grief, then, might help signal to others that one’s need is legitimate, and make one appear to be a better target of subsequent investment. The adaptive value of grief in this account lies not in what it makes the griever do per se; what the griever is doing is likely maladaptive in and of itself. However, that personally-maladaptive behavior can have an effect on others, leading them to provide benefits to the grieving individuals in an adaptive fashion. In other words, grief doesn’t serve to recalibrate the griever’s behavior so much as it serves to recalibrate the behavior of social others who might invest in you.

Of Pathogens And Social Support

Though I’m usually consistent with updating about once a week, this last week and a half has found me out of sorts. Apparently, some infection managed to get the better of my body for a while, and most of the available time I had went into managing my sickness and taking care of the most important tasks. Unfortunately, that also meant taking time away from writing, but now that I’m back on my feet I would like to offer some reflections on that rather grueling experience. One rather interesting – or annoying, if you’re me – facet of this last infection was the level of emotional intensity I found myself experiencing: I felt as if I wanted to be around other people while I was sick, which is something of an unusual experience for me; I found myself experiencing a greater degree of empathy with other people’s experiences than usual; I also found myself feeling, for lack of a better word, lonely, and a bit on the anxious side. Being the psychologist that I am, I couldn’t help but wonder what the ultimate function of these emotional experiences was. They certainly seemed to be driving me towards spending time around other people, but why?

And don’t you dare tell me it’s because company is pleasant; we all know that’s a lie.

Specifically, my question was whether these feelings of wanting to spend more time around others were being driven primarily by some psychological mechanism of mine functioning in my own fitness interests, or whether they might have been being driven by whatever parasite had colonized parts of my body. A case could be made for either option, though the case for parasite manipulation is admittedly more speculative, so let’s start with the idea that my increased desire for human contact might have been the result of the proper functioning of my psychology. Though I do not have any research on hand that directly examines the link between sickness and the desire for social closeness with others, I happen to have what is, perhaps, the next best thing: a paper by Aaroe & Petersen (2013) examining what effects hunger has on people’s willingness to advocate for resource-sharing behavior. Since the underlying theory behind the sickness-induced emotionality on my part and the hunger-induced resource sharing are broadly similar, examining the latter can help us understand the former.

Aaroe & Petersen (2013) begin with a relatively basic suggestion: solving the problems of resource acquisition posed an adaptive problem to ancestral human populations. We all need caloric resources to build and maintain our bodies, as well as to do all the reproductively-useful things that organisms which move about their environment do. One way of solving this problem, of course, is to go out hunting or foraging for food oneself. However, this strategy can, at times, be unsuccessful. Every now and again, people will come home empty-handed and hungry. If one happens to be a member of social species, like us, that’s not the only game in town, though: if you’re particularly cunning, you can manipulate successful others into sharing some of their resources with you. Accordingly, Aaroe & Petersen (2013) further suggest that humans might have evolved some cognitive mechanisms that responds to bodily signals of energy scarcity by attempting to persuade others to share more. Specifically, if your blood glucose level is low, you might be inclined to advocate for social policies that encourage others to share their resources with you.

As an initial test of this idea, the researchers had 104 undergraduates fast for four hours prior to the experiment. As if not eating for 4 hours wasn’t already a lot to ask. upon their arrival at the experiment, all the participants had their blood glucose levels measured in a process I can only assume (unfortunately for them) involved a needle. After the initial measurement, half the subjects were either given a sugar-rich drink (Spite) or a sugarless drink (Sprite Zero). Ten minutes after the drink, the blood glucose levels were measured again (and a third time as they leaving, which is a lot of pokes), and participants were asked about their support for various social redistribution policies. They were also asked to play a dictator game and divide approximately $350 between them and another participant, with one set of participants actually getting the money in that division. So the first test was designed to see whether participants would advocate for more sharing behavior when they were hungry, whereas the second test was designed to see if participants would actually demonstrate more generous behavior themselves.

Way to really earn your required undergrad research credits.

The results showed that the participants who had consumed the sugar-rich drink had higher blood glucose levels than the control group, and were also approximately 10% less supportive of social-welfare policies than those in the sugar-free condition. This lends some support to the idea that our current hunger level, at least as assessed by blood glucose levels, helps determine how much we are willing to advocate that other people share with one another: hungry individuals wanted more sharing, whereas less-hungry individuals wanted less. What about their actual sharing behavior, though? As it turns out, those who support social-welfare policies are more likely to share with others, but those who had low blood-glucose were less likely to do so. These two effects ended up washing out, with the result being that blood glucose had no effect on how much the participants actually decided to divide a potential resource themselves. While hungry individuals advocated that other people should share, then, they were no more likely to share themselves. They wanted others to be more generous without paying the costs of such generosity personally.

So perhaps my sickness-induced emotionality reflected something along those same lines: sick individuals find themselves unable to complete all sorts of tasks – such as resource acquisition or defense – as effectively as non-sick individuals. Our caloric resources are likely being devoted to other tasks, such as revving up our immune response. Thus, I might have desired that other people, in essence, take care of me while I was sick, with those emotions – such as increased loneliness or empathy – providing the proximate motivation to seek out such investment. If the current results are any indication, however, I would be unlikely to practice what I preach; I would want people to take care of me without my helping them anymore than usual. How very selfish of me and my emotions. So that covers the idea that my behavior was driven by some personal fitness benefits, but what about the alternative? The pathogens that were exploiting my body have their own set of fitness interests, after all, and part of those interests involves finding new hosts in which to exploit and reproduce. It follows, at least in theory, then, that the pathogens might be able to increase their own fitness by manipulating my mind in such a way so as to encourage me to seek out other conspecifics in my environment.

The more time I spent around others individuals, the greater the chance I would spread the infection, especially given how much I was coughing. If the pathogens affect my desire to be around others by making me feel lonely or anxious, then, they can increase their own fitness. This idea is by no means far-fetched. There are many known instances of pathogens influencing their host’s behavior, and I’ve written a little bit before about one of them: the psychological effects that malaria can have on the behavior of their host mosquitoes. Mosquitoes which are infected with malaria seem to preferentially feed from humans, whereas mosquitoes not so infected do not show any evidence of such preferential behavior. This likely owes to the malaria benefiting itself by manipulating the behavior of their mosquito host. The malaria wants to get from human to human, but it needs to do so via mosquito bites. If the malaria can make their host preferentially try and feed from humans, the malaria can reproduce quicker and more effectively. There are also some plausible theoretical reasons for suspecting that some pathogen(s) might play a role in the maintenance of human homosexual orientations, at least in males. The idea that pathogens can affect our psychologies more generally, then, is far from an impossibility.

“We hope you don’t mind us making your life miserable for this next week too much, because we’re doing it anyway.”

The question of interest, however, is whether the pathogens were responsible for my behavior directly or not. As promised, I don’t have an answer to the question. I don’t know what I was infected with specifically, much less what compounds it was or wasn’t releasing into my body, or what effect they might have had on my behavior. Further, if I already possessed some adaptions for seeking out social support when sick, there would be less of a selective pressure for the pathogens to encourage my doing so; I would already be spreading the pathogen incidentally through my behavior. The real point of this question is not to necessarily answer it, however, as much as it’s to get us thinking about how our psychology might not, at least at times, be our own, so to speak. There are countless other organisms living within (and outside of) our bodies that have their own sets of fitness interests which they might prefer we indulge, even at the expense of our own. As for me, I’m just happy to be healthy again, and to feel like my head is screwing back on to where it used to be.

References: Aaroe, L. & Petersen, M. (2013). Hunger games: Fluctuations in blood glucose levels influence support for social welfare. Psychological Science, 24, 2550-2556.

Why Parents Affect Children Less Than Many People Assume

Despite what a small handful of detractors have had to say, inclusive fitness theory has proved to be one of most valuable ideas we have for understanding much of the altruism we observe in both human and non-human species. The basic logic of inclusive fitness theory is simple: genes can increase their reproductive fitness by benefiting other bodies that contain copies of them. So, since you happen to share 50% of your genes in common by descent with a full sibling, you can, to some extent, increase your own reproductive fitness by increasing theirs. This logic is captured by the deceptively-tiny formula of rb > c. In English, rather than math, the formula states that altruism will be favored so long as the benefit delivered to the receiver, discounted by the degree of relatedness between the two, is greater than the cost to the giver. To use the sibling example again, altruism would be favored by selection if the the benefit you provided to a full sibling increased their reproductive success by twice as much (or more) than it cost you to give even if there was zero reciprocation.

“You scratch my back, and then you scratch my back again”

While this equation highlights why a lot of “good/nice” behaviors are observed – like childcare – there’s a darker side to this equation as well. By dividing each side of the inclusive fitness equation by r, you get this: b > c/r. What this new equation highlights is the selfish nature of these interactions: relatives can be selected to benefit themselves by inflicting costs on their kin. In the case of full siblings, I should be expected to value my benefiting twice as much, relative to theirs; for half siblings, I should value myself four-times as much, and so on. Let’s stick to full-siblings for now, just to stay consistent. Each sibling within a family should, all else being equal, be expected to value itself twice as much as they value any other sibling. The parents of these siblings, however, see things very differently: from the perspective of the parent, each of these siblings is equally related to them, so, in theory, they should value each of these offspring equally (again, all else being equal. All else is almost never equal, but let’s assume it is to keep the math easy).

This means that parents should prefer that their children act in a particular way: specifically, parents should prefer their children to help each other when the benefit to one outweighs the cost to the other, or b > c. The children, on the other hand, should only wish to behave that way when the benefit to their sibling is twice the cost of themselves, or 2b > c. This yields the following conclusion: how parents would like their children to behave does not necessarily correspond to what is in the child’s best fitness interests. Parents hoping to maximize their own fitness have different best interests from the children hoping to maximize theirs. Children who behave as their parents would prefer would be at a reproductive disadvantage, then, relative to children who were resistant to such parental expectations. This insight was formalized by Trivers (1974) when he wrote:

  “…an important feature of the argument presented here is that offspring cannot rely on parents for disinterested guidance. One expects the offspring to be pre-programmed to resist some parental teachings while being open to other forms. This is particularly true, as argued below, for parental teachings that affects the altruistic and egoistic tendencies of the offspring.” (p. 258)

While parents might feel as if they only acting in the best interests of their children, the logic of inclusive fitness suggests strongly that this feeling might represent an attempt at manipulating others, rather than a statement of fact. To avoid the risk of sounding one-sided, this argument cuts in the other direction as well: children might experience their parent’s treatment of them as being less-fair than it actually is, as each child would like to receive twice the investment that parents should be willing to give naturally. The take-home message of this point, however, is simply that children who were readily molded by their parents should be expected to have reproduced those tendencies less, relative to children who were not so affected. In some regards, children should be expected to actively disregard what their parents want for them.

“My parents want me to brush my teeth. They’re such fascists sometimes.”

There are other reasons to expect that parents should not tend to leave lasting impressions on their children’s eventual personalities. One of those very good reasons also has to do with the inclusive fitness logic laid out initially: because parents tend to be 50% genetically related to their children, parents should be expected to invest in their children fairly heavily, relative to non-children at least. The corollary to this idea is that non-parents of the child should be expected to treat them substantially different than their parents do. This means that a child should be relatively unable to learn what counts as appropriate behavior towards others more generally from their interactions with their parents. Just because a proud parent has hung their child’s scribbled artwork on the household refrigerator, it doesn’t mean that anyone else will come to think of the child as a great artist. A relationship with your parents is different than a relationship with your friends which is different from a sexual relationship in a great many ways. Even within these broad classes of relationships, you might behave differently with one friend than you do with another.

We should expect our behavior around these different individuals to be context-specific. What you learn about one relationship might not readily transfer to any other. Though a child might be unable to physically dominate their parents, they might be able to dominate their peers; some jokes might be appropriate amongst friends, but not with your boss. Though some of what you learn about how to behave around your parents might transfer to other situations (such as the language you speak, if your parents happen to speakers of the native tongue), it also may not. When it does not transfer, we should expect children to discard what they learned about how to behave around their parents in favor of more context-appropriate behaviors (indeed, when children find their parents speak a different language than their peers, the child will predominately learn to speak as their peers do; not their parents). While a parent’s behavior should be expected to influence how that child behaves around that parent, we should not necessarily expect it to influence the child’s behavior around anyone else.

It should come as little surprise, then, that being raised by the same parents doesn’t actually tend to make children any more similar with respect to their personality than being raised by different ones. Tellegan et al (1988) compared 44 identical twin (MZ) pairs raised apart with 217 identical twins reared together, along with 27 fraternal twins (DZ) reared apart and 114 reared together. In terms of their personality measures, the MZ twins were far more alike than the DZ twins,  as one would expect from their shared genetics. When it came to the personality measures, however, MZ twins reared together were more highly correlated on 7 of the measures, while those reared apart were more highly correlated on 6 of them. In terms of the DZ twins, those reared together were higher on 9 of the variables, whereas those reared apart were higher on the remaining 5. The size of these differences when they did exist was often exceedingly small, typically amounting to a correlation difference of about 0.1 between the pairs, or 1% of the variance.

Pick the one you want to keep. I’d recommend the cuter one.

Even if twins reared together ended up being substantially more similar than twins reared apart – which they didn’t – this would still not demonstrate that parenting was the cause of that similarity. After all, twins reared together tend to share more than their parents; they also tend to share various aspects of their wider social life, such as extended families, peer groups, and other social settings. There are good empirical and theoretical reasons for thinking that parents have less of a lasting effect on their children than many often suppose. That’s not to say that parents don’t have any effects on their children, mind you; just that the effects that they have ought to be largely limited to their particular relationship with the child in question, barring the infliction of any serious injuries or other such issues that will transfer from one context to another. Parents can certainly make their children more or less happy when they’re in each others presence, but so can friends and more intimate partners. In terms of shaping their children’s later personality, it truly takes a village.

References: Tellegen et al. (1988). Personality similarity in twins reared apart and together. Journal of Personality and Social Psychology, 54, 1031-1039.

Trivers, R. (1974). Parent-Offspring conflict. American Zoologist, 14, 249-264.

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.