No, it’s not The Incentives—it’s you

There’s a narrative I find kind of troubling, but that unfortunately seems to be growing more common in science. The core idea is that the mere existence of perverse incentives is a valid and sufficient reason to knowingly behave in an antisocial way, just as long as one first acknowledges the existence of those perverse incentives. The way this dynamic usually unfolds is that someone points out some fairly serious problem with the way many scientists behave—say, our collective propensity to p-hack as if it’s going out of style, or the fact that we insist on submitting our manuscripts to publishers that are actively trying to undermine our interests—and then someone else will say, “I know, right—but what are you going to do, those are the incentives.”

As best I can tell, the words “it’s the incentives” are magic. Once they’re uttered by someone, natural law demands that everyone else involved in the conversation immediately stop whatever else they were doing, solemnly nod, and mumble something to the effect that, yes, the incentives are very bad, very bad indeed, and it’s a real tragedy that so many smart, hard-working people are being crushed under the merciless, gigantic boot of The System. Then there’s usually a brief pause, and after that, everyone goes back to discussing whatever they were talking about a moment earlier.

Perhaps I’m getting senile in my early middle age, but my anecdotal perception is that it used to be that, when somebody pointed out to a researcher that they might be doing something questionable, that researcher would typically either (a) argue that they weren’t doing anything questionable (often incorrectly, because there used to be much less appreciation for some of the statistical issues involved), or (b) look uncomfortable for a little while, allow an awkward silence to bloom, and then change the subject. In the last few years, I’ve noticed that uncomfortable discussions about questionable practices disproportionately seem to end with a chuckle or shrug, followed by a comment to the effect that we are all extremely sophisticated human beings who recognize the complexity of the world we live in, and sure it would be great if we lived in a world where one didn’t have to occasionally engage in shenanigans, but that would be extremely naive, and after all, we are not naive, are we?

There is, of course,  an element of truth to this kind of response. I’m not denying that perverse incentives exist; they obviously do. There’s no question that many aspects of modern scientific culture systematically incentivize antisocial behavior, and I don’t think we can or should pretend otherwise. What I do object to quite strongly is the narrative that scientists are somehow helpless in the face of all these awful incentives—that we can’t possibly be expected to take any course of action that has any potential, however small, to impede our own career development.

“I would publish in open access journals,” your friendly neighborhood scientist will say. “But those have a lower impact factor, and I’m up for tenure in three years.”

Or: “if I corrected for multiple comparisons in this situation, my effect would go away, and then the reviewers would reject the paper.”

Or: “I can’t ask my graduate students to collect an adequately-powered replication sample; they need to publish papers as quickly as they can so that they can get a job.”

There are innumerable examples of this kind, and they’ve become so routine that it appears many scientists have stopped thinking about what the words they’re saying actually mean, and instead simply glaze over and nod sagely whenever the dreaded Incentives are invoked.

A random bystander who happened to eavesdrop on a conversation between a group of scientists kvetching about The Incentives could be forgiven for thinking that maybe, just maybe, a bunch of very industrious people who generally pride themselves on their creativity, persistence, and intelligence could find some way to work around, or through, the problem. And I think they would be right. The fact that we collectively don’t see it as a colossal moral failing that we haven’t figured out a way to get our work done without having to routinely cut corners in the rush for fame and fortune is deeply troubling.

It’s also aggravating on an intellectual level, because the argument that we’re all being egregiously and continuously screwed over by The Incentives is just not that good. I think there are a lot of reasons why researchers should be very hesitant to invoke The Incentives as a justification for why any of us behave the way we do. I’ll give nine of them here, but I imagine there are probably others.

1. You can excuse anything by appealing to The Incentives

No, seriously—anything. Once you start crying that The System is Broken in order to excuse your actions (or inactions), you can absolve yourself of responsibility for all kinds of behaviors that, on paper, should raise red flags. Consider just a few behaviors that few scientists would condone:

  • Fabricating data or results
  • Regulary threatening to fire trainees in order to scare them into working harder
  • Deliberately sabotaging competitors’ papers or grants by reviewing them negatively

I think it’s safe to say most of us consider such practices to be thoroughly immoral, yet there are obviously people who engage in each of them. And when those people are caught or confronted, one of the most common justifications they fall back on is… you guessed it: The Incentives! When Diederik Stapel confessed to fabricating the data used in over 50 publications, he didn’t explain his actions by saying “oh, you know, I’m probably a bit of a psychopath”; instead, he placed much of the blame squarely on The Incentives:

I did not withstand the pressure to score, to publish, the pressure to get better in time. I wanted too much, too fast. In a system where there are few checks and balances, where people work alone, I took the wrong turn. I want to emphasize that the mistakes that I made were not born out of selfish ends.

Stapel wasn’t acting selfishly, you see… he was just subject to intense pressures. Or, you know, Incentives.

Or consider these quotes from a New York Times article describing Stapel’s unraveling:

In his early years of research — when he supposedly collected real experimental data — Stapel wrote papers laying out complicated and messy relationships between multiple variables. He soon realized that journal editors preferred simplicity. “They are actually telling you: ‘Leave out this stuff. Make it simpler,'” Stapel told me. Before long, he was striving to write elegant articles.

The experiment — and others like it — didn’t give Stapel the desired results, he said. He had the choice of abandoning the work or redoing the experiment. But he had already spent a lot of time on the research and was convinced his hypothesis was valid. “I said — you know what, I am going to create the data set,” he told me.

Reading through such accounts, it’s hard to avoid the conclusion that Stapel’s self-narrative is strikingly similar to the one that gets tossed out all the time on social media, or in conference bar conversations: here I am, a good scientist trying to do an honest job, and yet all around me is a system that incentivizes deception and corner-cutting. What do you expect me to do?.

Curiously, I’ve never heard any of my peers—including many of the same people who are quick to invoke The Incentives to excuse their own imperfections—seriously endorse The Incentives as an acceptable justification for Stapel’s behavior. In Stapel’s case, the inference we overwhelmingly jump to is that there must be something deeply wrong with Stapel, seeing as the rest of us also face the same perverse incentives on a daily basis, yet we somehow manage to get by without fabricating data. But this conclusion should make us a bit uneasy, I think, because if it’s correct (and I think it is), it implies that we aren’t really such slaves to The Incentives after all. When our morals get in the way, we appear to be perfectly capable of resisting temptation. And I mean, it’s not even like it’s particularly difficult; I doubt many researchers actively have to fight the impulse to manipulate their data, despite the enormous incentives to do so. I submit that the reason many of us feel okay doing things like reporting exploratory results as confirmatory results, or failing to mention that we ran six other studies we didn’t report, is not really that The Incentives are forcing us to do things we don’t like, but that it’s easier to attribute our unsavory behaviors to unstoppable external forces than to take responsibility for them and accept the consequences.

Needless to say, I think this kind of attitude is fundamentally hypocritical. If we’re not comfortable with pariahs like Stapel blaming The Incentives for causing them to fabricate data, we shouldn’t use The Incentives as an excuse for doing things that are on the same spectrum, albeit less severe. If you think that what the words “I did not withstand the pressure to score” really mean when they fall out of Stapel’s mouth is something like “I’m basically a weak person who finds the thought of not being important so intolerable I’m willing to cheat to get ahead”, then you shouldn’t give yourself a free pass just because when you use that excuse, you’re talking about much smaller infractions. Consider the possibility that maybe, just like Stapel, you’re actually appealing to The Incentives as a crutch to avoid having to make your life very slightly more difficult.

2. It would break the world if everyone did it

When people start routinely accepting that The System is Broken and The Incentives Are Fucking Us Over, bad things tend to happen. It’s very hard to have a stable, smoothly functioning society once everyone believes (rightly or wrongly) that gaming the system is the only way to get by. Imagine if every time you went to your doctor—and I’m aware that this analogy won’t work well for people living outside the United States—she sent you to get a dozen expensive and completely unnecessary medical tests, and then, when prompted for an explanation, simply shrugged and said “I know I’m not an angel—but hey, them’s The Incentives.” You would be livid—even though it’s entirely true (at least in the United States; other developed countries seem to have figured this particular problem out) that many doctors have financial incentives to order unnecessary tests.

To be clear, I’m not saying perverse incentives never induce bad behavior in medicine or other fields. Of course they do. My point is that practitioners in other fields at least appear to have enough sense not to loudly trumpet The Incentives as a reasonable justification for their antisocial behavior—or to pat themselves on the back for being the kind of people who are clever enough to see the fiendish Incentives for exactly what they are. My sense is that when doctors, lawyers, journalists, etc. fall prey to The Incentives, they generally consider that to be a source of shame. I won’t go so far as to suggest that we scientists take pride in behaving badly—we obviously don’t—but we do seem to have collectively developed a rather powerful form of learned helplessness that doesn’t seem to be matched by other communities. Which is a fortunate thing, because if every other community also developed the same attitude, we would be in a world of trouble.

3. You are not special

Individual success in science is, to a first approximation, a zero-sum game—at least in the short term. many scientists who appeal to The Incentives seem to genuinely believe that opting out of doing the right thing is a victimless crime. I mean, sure, it might make the system a bit less efficient overall… but that’s just life, right? It’s not like anybody’s actually suffering.

Well yeah, people actually do suffer. There are many scientists who are willing to do the right things—to preregister their analysis plans, to work hard to falsify rather than confirm their hypotheses, to diligently draw attention to potential confounds that complicate their preferred story, and so on. When you assert your right to opt out of these things because apparently your publications, your promotions, and your students are so much more important than everyone else’s, you’re cheating those people.

No, really, you are. If you don’t like to think of yourself as someone who cheats other people, don’t reflexively collapse on a crutch made out of stainless steel Incentives any time someone questions your process. You are not special. Your publications, job, and tenure are not more important than other people’s. The fact that there are other people in your position engaging in the same behaviors doesn’t mean you and your co-authors are all very sophisticated, and that the people who refuse to cut corners are naive simpletons. What it actually demonstrates is that, somewhere along the way, you developed the reflexive ability to rationalize away behavior that you would disapprove of in others and that, viewed dispassionately, is clearly damaging to science.

4. You (probably) have no data

It’s telling that appeals to The Incentives are rarely supported by any actual data. It’s simply taken for granted that engaging in the practice in question would be detrimental to one’s career. The next time you’re tempted to blame The System for making you do bad things, you might want to ask yourself this: Do you actually know that, say, publishing in PLOS ONE rather than [insert closed society journal of your choice] would hurt your career? If so, how do you know that? Do you have any good evidence for it, or have you simply accepted it as stylized fact?

Coming by the kind of data you’d need to answer this question is actually not that easy: it’s not enough to reflexively point to, say, the fact that some journals have higher impact factors than others, To identify the utility-maximizing course of action, you’d need to integrate over both benefits and costs, and the costs are not always so obvious. For example, the opportunity cost of submitting your paper to a “good” journal will be offset to some extent by the likelihood of faster publication (no need to spend two years racking up rejections at high-impact venues), by the positive image you send to at least some of your peers that you support open scientific practices, and so on.

I’m not saying that a careful consideration of the pros and cons of doing the right thing would usually lead people to change their minds. It often won’t. What I’m saying is that people who blame The Incentives for forcing them to submit their papers to certain journals, to tell post-hoc stories about their work, or to use suboptimal analytical methods don’t generally support their decisions with data, or even with well-reasoned argument. The defense is usually completely reflexive—which should raise our suspicion that it’s also just a self-serving excuse.

5. It (probably) won’t matter anyway

This one might hurt a bit, but I think it’s important to consider—particularly for early-career researchers. Let’s suppose you’re right that doing the right thing in some particular case would hurt your career. Maybe it really is true that if you comprehensively report in your paper on all the studies you ran, and not just the ones that “worked”, your colleagues will receive your work less favorably. In such cases it may seem natural to think that there has to be a tight relationship between the current decision and the global outcome—i.e., that if you don’t drop the failed studies, you won’t get a tenure-track position three years down the road. After all, you’re focusing on that causal relationship right now, and it seems so clear in your head!

Unfortunately (or perhaps fortunately?), reality doesn’t operate that way. Outcomes in academia are multiply determined and enormously complex. You can tell yourself that getting more papers out faster will get you a job if it makes you feel better, but that doesn’t make it true. If you’re a graduate student on the job market these days, I have sad news for you: you’re probably not getting a tenure-track job no matter what you do. It doesn’t matter how many p-hacked papers you publish, or how thinly you slice your dissertation into different “studies”; there are not nearly enough jobs to go around for everyone who wants one.

Suppose you’re right, and your sustained pattern of corner-cutting is in fact helping you get ahead. How far ahead do you think it’s helping you get? Is it taking you from a 3% chance of getting a tenure-track position at an R1 university to an 80% chance? Almost certainly not. Maybe it’s increasing that probability from 7% to 11%; that would still be a non-trivial relative increase, but it doesn’t change the fact that, for the average grad student, there is no full-time faculty position waiting at the end of the road. Despite what the environment around you may make you think, the choice most graduate students and postdocs face is not actually between (a) maintaining your integrity and “failing” out of science or (b) cutting a few corners and achieving great fame and fortune as a tenured professor. The Incentives are just not that powerful. The vastly more common choice you face as a trainee is between (a) maintaining your integrity and having a pretty low chance of landing a permanent research position, or (b) cutting a bunch of corners that threaten the validity of your work and having a slightly higher (but still low in absolute terms) chance of landing a permanent research position. And even that’s hardly guaranteed, because you never know when there’s someone on a hiring committee who’s going to be turned off by the obvious p-hacking in your work.

The point is, the world is complicated, and as a general rule, very few things—including the number of publications you produce—are as important as they seem to be when you’re focusing on them in the moment. If you’re an early-career researcher and you regularly find yourself strugging between doing what’s right and doing what isn’t right but (you think) benefits your career, you may want to take a step back and dispassionately ask yourself whether this integrity versus expediency conflict is actually a productive way to frame things. Instead, consider the alternative framing I suggested above: you are most likely going to leave academia eventually, no matter what you do, so why not at least try to see the process through with some intellectual integrity? And I mean, if you’re really so convinced that The System is Broken, why would you want to stay in it anyway? Do you think standards are going to change dramatically in the next few years? Are you laboring under the impression that you, of all people, are going to somehow save science?

This brings us directly to the next point…

6. You’re (probably) not going to “change things from the inside”

Over the years, I’ve talked to quite a few early-career researchers who have told me that while they can’t really stop engaging in questionable research practices right now without hurting their career, they’re definitely going to do better once they’re in a more established position. These are almost invariably nice, well-intentioned people, and I don’t doubt that they genuinely believe what they say. Unfortunately, what they say is slippery, and has a habit of adapting to changing circumstances. As a grad student or postdoc, it’s easy to think that once you get a faculty position, you’ll be able to start doing research the “right” way. But once you get a faculty position, it then turns out you need to get papers and grants in order to get tenure (I mean, who knew?), so you decide to let the dreaded Incentives win for just a few more years. And then, once you secure tenure, well, now the problem is that your graduate students also need jobs, just like you once did, so you can’t exactly stop publishing at the same rate, can you? Plus, what would all your colleagues think if you effectively said, “oh, you should all treat the last 15 years of my work with skepticism—that was just for tenure”?

I’m not saying there aren’t exceptions. I’m sure there are. But I can think of at least a half-dozen people off-hand who’ve regaled me with me some flavor of “once I’m in a better position” story, and none of them, to my knowledge, have carried through on their stated intentions in a meaningful way. And I don’t find this surprising: in most walks of life, course correction generally becomes harder, not easier, the longer you’ve been traveling on the wrong bearing. So if part of your unhealthy respect for The Incentives is rooted in an expectation that those Incentives will surely weaken their grip on you just as soon as you reach the next stage of your career, you may want to rethink your strategy. The Incentives are not going to dissipate as you move up the career ladder; if anything, you’re probably going to have an increasingly difficult time shrugging them off.

7. You’re not thinking long-term

One of the most frustrating aspects of appeals to The Incentives is that they almost invariably seem to focus exclusively on the short-to-medium term. But the long term also matters. And there, I would argue that The Incentives very much favor a radically different—and more honest—approach to scientific research. To see this, we need only consider the ongoing “replication crisis” in many fields of science. One thing that I think has been largely overlooked in discussions about the current incentive structure of science is what impact the replication crisis will have on the legacies of a huge number of presently famous scientists.

I’ll tell you what impact it will have: many of those legacies will be completely zeroed out. And this isn’t just hypothetical scaremongering. It’s happening right now to many former stars of psychology (and, I imagine, other fields I’m less familiar with). There are many researchers we can point to right now who used to be really famous (like, major-chunks-of-the-textbook famous), are currently famous-with-an-asterisk, and will in all likelihood, be completely unknown again within a couple of decades. The unlucky ones are probably even fated to become infamous—their entire scientific legacies eventually reduced to footnotes in cautionary histories illustrating how easily entire areas of scientific research can lose their footing when practitioners allow themselves to be swept away by concerns about The Incentives.

You probably don’t want this kind of thing to happen to you. I’m guessing you would like to retire with at least some level of confidence that your work, while maybe not Earth-shattering in its implications, isn’t going to be tossed on the scrap heap of history one day by a new generation of researchers amazed at how cavalier you and your colleagues once were about silly little things like “inferential statistics” and “accurate reporting”. So if your justification for cutting corners is that you can’t otherwise survive or thrive in the present environment, you should consider the prospect—and I mean, really take some time to think about it—that any success you earn within the next 10 years by playing along with The Incentives could ultimately make your work a professional joke within the 20 years after that.

8. It achieves nothing and probably makes things worse

Hey, are you a scientist? Yes? Great, here’s a quick question for you: do you think there’s any working scientist on Planet Earth who doesn’t already know that The Incentives are fucked up? No? I didn’t think so. Which means you really don’t need to keep bemoaning The Incentives; I promise you that you’re not helping to draw much-needed attention to an important new problem nobody’s recognized before. You’re not expressing any deep insight by pointing out that hiring committees prefer applicants with lots of publications in high-impact journals to applicants with a few publications in journals no one’s ever heard of. If your complaints are achieving anything at all, they’re probably actually making things worse by constantly (and incorrectly) reminding everyone around you about just how powerful The Incentives are.

Here’s a suggestion: maybe try not talking about The Incentives for a while. You could even try, I don’t know, working against The Incentives for a change. Or, if you can’t do that, just don’t say anything at all. Probably nobody will miss anything, and the early-career researchers among us might even be grateful for a respite from their senior colleagues’ constant reminder that The System—the very same system those senior colleagues are responsible for creating!—is so fucked up.

9. It’s your job

This last one seems so obvious it should go without saying, but it does need saying, so I’ll say it: a good reason why you should avoid hanging bad behavior on The Incentives is that you’re a scientist, and trying to get closer to the truth, and not just to tenure, is in your fucking job description. Taxpayers don’t fund you because they care about your career; they fund you to learn shit, cure shit, and build shit. If you can’t do your job without having to regularly excuse sloppiness on the grounds that you have no incentive to be less sloppy, at least have the decency not to say that out loud in a crowded room or Twitter feed full of people who indirectly pay your salary. Complaining that you would surely do the right thing if only these terrible Incentives didn’t exist doesn’t make you the noble martyr you think it does; to almost anybody outside your field who has a modicum of integrity, it just makes you sound like you’re looking for an easy out. It’s not sophisticated or worldly or politically astute, it’s just dishonest and lazy. If you find yourself unable to do your job without regularly engaging in practices that clearly devalue the very science you claim to care about, and this doesn’t bother you deeply, then maybe the problem is not actually The Incentives—or at least, not The Incentives alone. Maybe the problem is You.

In defense of In Defense of Facebook

A long, long time ago (in social media terms), I wrote a post defending Facebook against accusations of ethical misconduct related to a newly-published study in PNAS. I won’t rehash the study, or the accusations, or my comments in any detail here; for that, you can read the original post (I also recommend reading this or this for added context). While I stand by most of what I wrote, as is the nature of things, sometimes new information comes to light, and sometimes people say things that make me change my mind. So I thought I’d post my updated thoughts and reactions. I also left some additional thoughts in a comment on my last post, which I won’t rehash here.

Anyway, in no particular order…

I’m not arguing for a lawless world where companies can do as they like with your data

Some people apparently interpreted my last post as a defense of Facebook’s data use policy in general. It wasn’t. I probably brought this on myself in part by titling the post “In Defense of Facebook”. Maybe I should have called it something like “In Defense of this one particular study done by one Facebook employee”. In any case, I’ll reiterate: I’m categorically not saying that Facebook–or any other company, for that matter–should be allowed to do whatever it likes with its users’ data. There are plenty of valid concerns one could raise about the way companies like Facebook store, manage, and use their users’ data. And for what it’s worth, I’m generally in favor of passing new rules regulating the use of personal data in the private sector. So, contrary to what some posts suggested, I was categorically not advocating for a laissez-faire world in which large corporations get to do as they please with your information, and there’s nothing us little people can do about it.

The point I made in my last post was much narrower than that–namely, that picking on the PNAS study as an example of ethically questionable practices at Facebook was a bad idea, because (a) there aren’t any new risks introduced by this manipulation that aren’t already dwarfed by the risks associated with using Facebook itself (which is not exactly a high-risk enterprise to begin with), and (b) there are literally thousands of experiments just like this being conducted every day by large companies intent on figuring out how best to market their products and services–so Facebook’s study doesn’t stand out in any respect. My point was not that you shouldn’t be concerned about who has your data and how they’re using it, but that it’s deeply counterproductive to go after Facebook for this particular experiment when Facebook is of the few companies in this arena who actually (occasionally) publish the results of their findings in the scientific literature, instead of hiding them entirely from the light, as almost everyone else does. Of course, that will probably change as a result of this controversy.

I Was Wrong–A/B Testing Edition.

One claim I made in my last post that was very clearly wrong is this (emphasis added):

What makes the backlash on this issue particularly strange is that I’m pretty sure most people do actually realize that their experience on Facebook (and on other websites, and on TV, and in restaurants, and in museums, and pretty much everywhere else) is constantly being manipulated. I expect that most of the people who’ve been complaining about the Facebook study on Twitter are perfectly well aware that Facebook constantly alters its user experience–I mean, they even see it happen in a noticeable way once in a while, whenever Facebook introduces a new interface.

After watching the commentary over the past two days, I think it’s pretty clear I was wrong about this. A surprisingly large number of people clearly were genuinely unaware that Facebook, Twitter, Google, and other major players in every major industry (not just tech–also banks, groceries, department stores, you name it) are constantly running large-scale, controlled experiments on their users and customers. For instance, here’s a telling comment left on my last post:

The main issue I have with the experiment is that they conducted it without telling us. Given, that would have been counterproductive, but even a small adverse affect is still an adverse affect. I just don’t like the idea that corporations can do stuff to me without my consent. Just my opinion.

Similar sentiments are all over the place. Clearly, the revelation that Facebook regularly experiments on its users without their knowledge was indeed just that to many people–a revelation. I suppose in this sense, there’s potentially a considerable upside to this controversy, inasmuch as it has clearly served to raise awareness of industry-standard practices.

Questions about the ethics of the PNAS paper’s publication

My post focused largely on the question of whether the experiment Facebook conducted was itself illegal or unethical. I took this to be the primary concern of most lay people who have expressed concern about the episode. As I discussed in my post, I think it’s quite clear that the experiment itself is (a) entirely legal and that (b) any ethical objections one could raise are actually much broader objections about the way we regulate data use and consumer privacy, and have nothing to do with Facebook in particular. However, there’s a separate question that does specifically concern Facebook–or really, the authors of the PNAS paper–which is whether the authors, in their efforts to publish their findings, violated any laws or regulations.

When I wrote my post, I was under the impression–based largely on reports of an interview with the PNAS editor, Susan Fiske–that the authors had in fact obtained approval to conduct the study from an IRB, and had simply neglected to include that information in the text (which would have been an Editorial lapse, but not an unethical act). I wrote as much in a comment on my post. I was not suggesting–as some seemed to take away–that Facebook doesn’t need to get IRB approval. I was operating on the assumption that it had obtained IRB approval, based on the information available at the time.

In any case, it now appears that may not be exactly what happened. Unfortunately, it’s not yet clear exactly what did happen. One version of events people have suggested is that the study’s authors exploited a loophole in the rules by having Facebook conduct and analyze the experiment without the involvement of the other authors–who only contributed to the genesis of the idea and the writing of the manuscript. However, this interpretation is not unambiguous, and risks maligning the authors’ reputations unfairly, because Adam Kramer’s post explaining the motivation for the experiment suggests that the idea for the experiment originated entirely at Facebook, and was related to internal needs:

The reason we did this research is because we care about the emotional impact of Facebook and the people that use our product. We felt that it was important to investigate the common worry that seeing friends post positive content leads to people feeling negative or left out. At the same time, we were concerned that exposure to friends’ negativity might lead people to avoid visiting Facebook. We didn’t clearly state our motivations in the paper.

How you interpret the ethics of the study thus depends largely on what you believe actually happened. If you believe that the genesis and design of the experiment were driven by Facebook’s internal decision-making, and the decision to publish an interesting finding came only later, then there’s nothing at all ethically questionable about the authors’ behavior. It would have made no more sense to seek out IRB approval for this one experiment than for any of the other in-house experiments Facebook regularly conducts. And there is, again, no question whatsoever that Facebook does not have to get approval from anyone to do experiments that are not for the purpose of systematic, generalizable research.

Moreover, since the non-Facebook authors did in fact ask the IRB to review their proposal to use archival data–and the IRB exempted them from review, as is routinely done for this kind of analysis–there would be no legitimacy to the claim that the authors acted unethically. About the only claim one could raise an eyebrow at is that the authors “didn’t clearly state” their motivations. But since presenting a post-hoc justification for one’s studies that has nothing to do with the original intention is extremely common in psychology (though it shouldn’t be), it’s not really fair to fault Kramer et al for doing something that is standard practice.

If, on the other hand, the idea for the study did originate outside of Facebook, and the authors deliberately attempted to avoid prospective IRB review, then I think it’s fair to say that their behavior was unethical. However, given that the authors were following the letter of the law (if clearly not the spirit), it’s not clear that PNAS should have, or could have, rejected the paper. It certainly should have demanded that information regarding interactions with the IRB be included in the manuscript, and perhaps it could have published some kind of expression of concern alongside the paper. But I agree with Michelle Meyer’s analysis that, in taking the steps they took, the authors are almost certainly operating within the rules, because (a) Facebook itself is not subject to HHS rules, (b) the non-Facebook authors were not technically “engaged in research”, and (c) the archival use of already-collected data by the non-Facebook authors was approved by the Cornell IRB (or rather, the study was exempted from further review).

Absent clear evidence of what exactly happened in the lead-up to publication, I think the appropriate course of action is to withhold judgment. In the interim, what the episode clearly does do is lay bare how ill-prepared the existing HHS regulations are for dealing with the research use of data collected online–particularly when the data was acquired by private entities. Actually, it’s not just research use that’s problematic; it’s clear that many people complaining about Facebook’s conduct this week don’t really give a hoot about the “generalizable knowledge” side of things, and are fundamentally just upset that Facebook is allowed to run these kinds of experiments at all without providing any notification.

In my view, what’s desperately called for is a new set of regulations that provide a unitary code for dealing with consumer data across the board–i.e., in both research and non-research contexts. This leaves aside exactly what such regulations would look like, of course. My personal view is that the right direction to move in is to tighten consumer protection laws to better regulate management and use of private citizens’ data, while simultaneously liberalizing the research use of private datasets that have already been acquired. For example, I would favor a law that (a) forced Facebook and other companies to more clearly and explicitly state how they use their users’ data, (b) provided opt-out options when possible, along with the ability for users to obtain report of how their data has been used in the past, and (c) gave blanket approval to use data acquired under these conditions for any and all academic research purposes so long as the data are deidentified. Many people will disagree with this, of course, and have very different ideas. That’s fine; the key point is that the conversation we should be having is about how to update and revise the rules governing research vs. non-research uses of data in such a way that situations like the PNAS study don’t come up again.

What Facebook does is not research–until they try to publish it

Much of the outrage over the Facebook experiment is centered around the perception that Facebook shouldn’t be allowed to conduct research on its users without their consent. What many people mean by this, I think, is that Facebook shouldn’t be allowed to conduct any experiments on its users for purposes of learning things about user experience and behavior unless Facebook explicitly asks for permission. A point that I should have clarified in my original post is that Facebook users are, in the normal course of things, not considered participants in a research study, no matter how or how much their emotions are manipulated. That’s because the HHS’s definition of research includes, as a necessary component, that there be an active intention to contribute to generalizable new knowledge.

Now, to my mind, this isn’t a great way to define “research”–I think it’s a good idea to avoid definitions that depend on knowing what people’s intentions were when they did something. But that’s the definition we’re stuck with, and there’s really no ambiguity over whether Facebook’s normal operations–which include constant randomized, controlled experimentation on its users–constitute research in this sense. They clearly don’t. Put simply, if Facebook were to eschew disseminating its results to the broader community, the experiment in question would not have been subject to any HHS regulations whatsoever (though, as Michelle Meyer astutely pointed out, technically the experiment probably isn’t subject to HHS regulation even now, so the point is moot). Again, to reiterate: it’s only the fact that Kramer et al wanted to publish their results in a scientific journal that opened them up to criticism of research misconduct in the first place.

This observation may not have any impact on your view if your concern is fundamentally about the publication process–i.e., you don’t object to Facebook doing the experiment; what you object to is Facebook trying to disseminate their findings as research. But it should have a strong impact on your views if you were previously under the impression that Facebook’s actions must have violated some existing human subjects regulation or consumer protection law. The laws in the United States–at least as I understand them, and I admittedly am not a lawyer–currently afford you no such protection.

Now, is it a good idea to have two very separate standards, one for research and one for everything else? Probably not. Should Facebook be allowed to do whatever it wants to your user experience so long as it’s covered under the Data Use policy in the user agreement you didn’t read? Probably not. But what’s unequivocally true is that, as it stands right now, your interactions with Facebook–no matter how your user experience, data, or emotions are manipulated–are not considered research unless Facebook manipulates your experience with the express intent of disseminating new knowledge to the world.

Informed consent is not mandatory for research studies

As a last point, there seems to be a very common misconception floating around among commentators that the Facebook experiment was unethical because it didn’t provide informed consent, which is a requirement for all research studies involving experimental manipulation. I addressed this in the comments on my last post in response to other comments:

[I]t’s simply not correct to suggest that all human subjects research requires informed consent. At least in the US (where Facebook is based), the rules governing research explicitly provide for a waiver of informed consent. Directly from the HHS website:

An IRB may approve a consent procedure which does not include, or which alters, some or all of the elements of informed consent set forth in this section, or waive the requirements to obtain informed consent provided the IRB finds and documents that:

(1) The research involves no more than minimal risk to the subjects;

(2) The waiver or alteration will not adversely affect the rights and welfare of the subjects;

(3) The research could not practicably be carried out without the waiver or alteration; and

(4) Whenever appropriate, the subjects will be provided with additional pertinent information after participation.

Granting such waivers is a commonplace occurrence; I myself have had online studies granted waivers before for precisely these reasons. In this particular context, it’s very clear that conditions (1) and (2) are met (because this easily passes the “not different from ordinary experience” test). Further, Facebook can also clearly argue that (3) is met, because explicitly asking for informed consent is likely not viable given internal policy, and would in any case render the experimental manipulation highly suspect (because it would no longer be random). The only point one could conceivably raise questions about is (4), but here again I think there’s a very strong case to be made that Facebook is not about to start providing debriefing information to users every time it changes some aspect of the news feed in pursuit of research, considering that its users have already agreed to its User Agreement, which authorizes this and much more.

Now, if you disagree with the above analysis, that’s fine, but what should be clear enough is that there are many IRBs (and I’ve personally interacted with some of them) that would have authorized a waiver of consent in this particular case without blinking. So this is clearly well within “reasonable people can disagree” territory, rather than “oh my god, this is clearly illegal and unethical!” territory.

I can understand the objection that Facebook should have applied for IRB approval prior to conducting the experiment (though, as I note above, that’s only true if the experiment was initially conducted as research, which is not clear right now). However, it’s important to note that there is no guarantee that an IRB would have insisted on informed consent at all in this case. There’s considerable heterogeneity in different IRBs’ interpretation of the HHS guidelines (and in fact, even across different reviewers within the same IRB), and I don’t doubt that many IRBs would have allowed Facebook’s application to sail through without any problems (see, e.g., this comment on my last post)–though I think there’s a general consensus that a debriefing of some kind would almost certainly be requested.

In defense of Facebook

[UPDATE July 1st: I’ve now posted some additional thoughts in a second post here.]

It feels a bit strange to write this post’s title, because I don’t find myself defending Facebook very often. But there seems to be some discontent in the socialmediaverse at the moment over a new study in which Facebook data scientists conducted a large-scale–over half a million participants!–experimental manipulation on Facebook in order to show that emotional contagion occurs on social networks. The news that Facebook has been actively manipulating its users’ emotions has, apparently, enraged a lot of people.

The study

Before getting into the sources of that rage–and why I think it’s misplaced–though, it’s worth describing the study and its results. Here’s a description of the basic procedure, from the paper:

The experiment manipulated the extent to which people (N = 689,003) were exposed to emotional expressions in their News Feed. This tested whether exposure to emotions led people to change their own posting behaviors, in particular whether exposure to emotional content led people to post content that was consistent with the exposure—thereby testing whether exposure to verbal affective expressions leads to similar verbal expressions, a form of emotional contagion. People who viewed Facebook in English were qualified for selection into the experiment. Two parallel experiments were conducted for positive and negative emotion: One in which exposure to friends’ positive emotional content in their News Feed was reduced, and one in which exposure to negative emotional content in their News Feed was reduced. In these conditions, when a person loaded their News Feed, posts that contained emotional content of the relevant emotional valence, each emotional post had between a 10% and 90% chance (based on their User ID) of being omitted from their News Feed for that specific viewing.

And here’s their central finding:

What the figure shows is that, in the experimental conditions, where negative or positive emotional posts are censored, users produce correspondingly more positive or negative emotional words in their own status updates. Reducing the number of negative emotional posts users saw led those users to produce more positive, and fewer negative words (relative to the unmodified control condition); conversely, reducing the number of presented positive posts led users to produce more negative and fewer positive words of their own.

Taken at face value, these results are interesting and informative. For the sake of contextualizing the concerns I discuss below, though, two points are worth noting. First, these effects, while highly statistically significant, are tiny. The largest effect size reported had a Cohen’s d of 0.02–meaning that eliminating a substantial proportion of emotional content from a user’s feed had the monumental effect of shifting that user’s own emotional word use by two hundredths of a standard deviation. In other words, the manipulation had a negligible real-world impact on users’ behavior. To put it in intuitive terms, the effect of condition in the Facebook study is roughly comparable to a hypothetical treatment that increased the average height of the male population in the United States by about one twentieth of an inch (given a standard deviation of ~2.8 inches). Theoretically interesting, perhaps, but not very meaningful in practice.

Second, the fact that users in the experimental conditions produced content with very slightly more positive or negative emotional content doesn’t mean that those users actually felt any differently. It’s entirely possible–and I would argue, even probable–that much of the effect was driven by changes in the expression of ideas or feelings that were already on users’ minds. For example, suppose I log onto Facebook intending to write a status update to the effect that I had an “awesome day today at the beach with my besties!” Now imagine that, as soon as I log in, I see in my news feed that an acquaintance’s father just passed away. I might very well think twice about posting my own message–not necessarily because the news has made me feel sad myself, but because it surely seems a bit unseemly to celebrate one’s own good fortune around people who are currently grieving. I would argue that such subtle behavioral changes, while certainly responsive to others’ emotions, shouldn’t really be considered genuine cases of emotional contagion. Yet given how small the effects were, one wouldn’t need very many such changes to occur in order to produce the observed results. So, at the very least, the jury should still be out on the extent to which Facebook users actually feel differently as a result of this manipulation.

The concerns

Setting aside the rather modest (though still interesting!) results, let’s turn to look at the criticism. Here’s what Katy Waldman, writing in a Slate piece titled “Facebook’s Unethical Experiment“, had to say:

The researchers, who are affiliated with Facebook, Cornell, and the University of California–San Francisco, tested whether reducing the number of positive messages people saw made those people less likely to post positive content themselves. The same went for negative messages: Would scrubbing posts with sad or angry words from someone’s Facebook feed make that person write fewer gloomy updates?

The upshot? Yes, verily, social networks can propagate positive and negative feelings!

The other upshot: Facebook intentionally made thousands upon thousands of people sad.

Or consider an article in the The Wire, quoting Jacob Silverman:

“What’s disturbing about how Facebook went about this, though, is that they essentially manipulated the sentiments of hundreds of thousands of users without asking permission (blame the terms of service agreements we all opt into). This research may tell us something about online behavior, but it’s undoubtedly more useful for, and more revealing of, Facebook’s own practices.”

On Twitter, the reaction to the study has been similarly negative). A lot of people appear to be very upset at the revelation that Facebook would actively manipulate its users’ news feeds in a way that could potentially influence their emotions.

Why the concerns are misplaced

To my mind, the concerns expressed in the Slate piece and elsewhere are misplaced, for several reasons. First, they largely mischaracterize the study’s experimental procedures–to the point that I suspect most of the critics haven’t actually bothered to read the paper. In particular, the suggestion that Facebook “manipulated users’ emotions” is quite misleading. Framing it that way tacitly implies that Facebook must have done something specifically designed to induce a different emotional experience in its users. In reality, for users assigned to the experimental condition, Facebook simply removed a variable proportion of status messages that were automatically detected as containing positive or negative emotional words. Let me repeat that: Facebook removed emotional messages for some users. It did not, as many people seem to be assuming, add content specifically intended to induce specific emotions. Now, given that a large amount of content on Facebook is already highly emotional in nature–think about all the people sharing their news of births, deaths, break-ups, etc.–it seems very hard to argue that Facebook would have been introducing new risks to its users even if it had presented some of them with more emotional content. But it’s certainly not credible to suggest that replacing 10% – 90% of emotional content with neutral content constitutes a potentially dangerous manipulation of people’s subjective experience.

Second, it’s not clear what the notion that Facebook users’ experience is being “manipulated” really even means, because the Facebook news feed is, and has always been, a completely contrived environment. I hope that people who are concerned about Facebook “manipulating” user experience in support of research realize that Facebook is constantly manipulating its users’ experience. In fact, by definition, every single change Facebook makes to the site alters the user experience, since there simply isn’t any experience to be had on Facebook that isn’t entirely constructed by Facebook. When you log onto Facebook, you’re not seeing a comprehensive list of everything your friends are doing, nor are you seeing a completely random subset of events. In the former case, you would be overwhelmed with information, and in the latter case, you’d get bored of Facebook very quickly. Instead, what you’re presented with is a carefully curated experience that is, from the outset, crafted in such a way as to create a more engaging experience (read: keeps you spending more time on the site, and coming back more often). The items you get to see are determined by a complex and ever-changing algorithm that you make only a partial contribution to (by indicating what you like, what you want hidden, etc.). It has always been this way, and it’s not clear that it could be any other way. So I don’t really understand what people mean when they sarcastically suggest–as Katy Waldman does in her Slate piece–that “Facebook reserves the right to seriously bum you out by cutting all that is positive and beautiful from your news feed”. Where does Waldman think all that positive and beautiful stuff comes from in the first place? Does she think it spontaneously grows wild in her news feed, free from the meddling and unnatural influence of Facebook engineers?

Third, if you were to construct a scale of possible motives for manipulating users’ behavior–with the global betterment of society at one end, and something really bad at the other end–I submit that conducting basic scientific research would almost certainly be much closer to the former end than would the other standard motives we find on the web–like trying to get people to click on more ads. The reality is that Facebook–and virtually every other large company with a major web presence–is constantly conducting large controlled experiments on user behavior. Data scientists and user experience researchers at Facebook, Twitter, Google, etc. routinely run dozens, hundreds, or thousands of experiments a day, all of which involve random assignment of users to different conditions. Typically, these manipulations aren’t conducted in order to test basic questions about emotional contagion; they’re conducted with the explicit goal of helping to increase revenue. In other words, if the idea that Facebook would actively try to manipulate your behavior bothers you, you should probably stop reading this right now and go close your account. You also should definitely not read this paper suggesting that a single social message on Facebook prior to the last US presidential election the may have single-handedly increased national voter turn-out by as much as 0.6%). Oh, and you should probably also stop using Google, YouTube, Yahoo, Twitter, Amazon, and pretty much every other major website–because I can assure you that, in every single case, there are people out there who get paid a good salary to… yes, manipulate your emotions and behavior! For better or worse, this is the world we live in. If you don’t like it, you can abandon the internet, or at the very least close all of your social media accounts. But the suggestion that Facebook is doing something unethical simply by publishing the results of one particular experiment among thousands–and in this case, an experiment featuring a completely innocuous design that, if anything, is probably less motivated by a profit motive than most of what Facebook does–seems kind of absurd.

Fourth, it’s worth keeping in mind that there’s nothing intrinsically evil about the idea that large corporations might be trying to manipulate your experience and behavior. Everybody you interact with–including every one of your friends, family, and colleagues–is constantly trying to manipulate your behavior in various ways. Your mother wants you to eat more broccoli; your friends want you to come get smashed with them at a bar; your boss wants you to stay at work longer and take fewer breaks. We are always trying to get other people to feel, think, and do certain things that they would not otherwise have felt, thought, or done. So the meaningful question is not whether people are trying to manipulate your experience and behavior, but whether they’re trying to manipulate you in a way that aligns with or contradicts your own best interests. The mere fact that Facebook, Google, and Amazon run experiments intended to alter your emotional experience in a revenue-increasing way is not necessarily a bad thing if in the process of making more money off you, those companies also improve your quality of life. I’m not taking a stand one way or the other, mind you, but simply pointing out that without controlled experimentation, the user experience on Facebook, Google, Twitter, etc. would probably be very, very different–and most likely less pleasant. So before we lament the perceived loss of all those “positive and beautiful” items in our Facebook news feeds, we should probably remind ourselves that Facebook’s ability to identify and display those items consistently is itself in no small part a product of its continual effort to experimentally test its offering by, yes, experimentally manipulating its users’ feelings and thoughts.

What makes the backlash on this issue particularly strange is that I’m pretty sure most people do actually realize that their experience on Facebook (and on other websites, and on TV, and in restaurants, and in museums, and pretty much everywhere else) is constantly being manipulated. I expect that most of the people who’ve been complaining about the Facebook study on Twitter are perfectly well aware that Facebook constantly alters its user experience–I mean, they even see it happen in a noticeable way once in a while, whenever Facebook introduces a new interface. Given that Facebook has over half a billion users, it’s a foregone conclusion that every tiny change Facebook makes to the news feed or any other part of its websites induces a change in millions of people’s emotions. Yet nobody seems to complain about this much–presumably because, when you put it this way, it seems kind of silly to suggest that a company whose business model is predicated on getting its users to use its product more would do anything other than try to manipulate its users into, you know, using its product more.

Why the backlash is deeply counterproductive

Now, none of this is meant to suggest that there aren’t legitimate concerns one could raise about Facebook’s more general behavior–or about the immense and growing social and political influence that social media companies like Facebook wield. One can certainly question whether it’s really fair to expect users signing up for a service like Facebook’s to read and understand user agreements containing dozens of pages of dense legalese, or whether it would make sense to introduce new regulations on companies like Facebook to ensure that they don’t acquire or exert undue influence on their users’ behavior (though personally I think that would be unenforceable and kind of silly). So I’m certainly not suggesting that we give Facebook, or any other large web company, a free pass to do as it pleases. What I am suggesting, however, is that even if your real concerns are, at bottom, about the broader social and political context Facebook operates in, using this particular study as a lightning rod for criticism of Facebook is an extremely counterproductive, and potentially very damaging, strategy.

Consider: by far the most likely outcome of the backlash Facebook is currently experiencing is that, in future, its leadership will be less likely to allow its data scientists to publish their findings in the scientific literature. Remember, Facebook is not a research institute expressly designed to further understanding of the human condition; it’s a publicly-traded corporation that exists to create wealth for its shareholders. Facebook doesn’t have to share any of its data or findings with the rest of the world if it doesn’t want to; it could comfortably hoard all of its knowledge and use it for its own ends, and no one else would ever be any wiser for it. The fact that Facebook is willing to allow its data science team to spend at least some of its time publishing basic scientific research that draws on Facebook’s unparalleled resources is something to be commended, not criticized.

There is little doubt that the present backlash will do absolutely nothing to deter Facebook from actually conducting controlled experiments on its users, because A/B testing is a central component of pretty much every major web company’s business strategy at this point–and frankly, Facebook would be crazy not to try to empirically determine how to improve user experience. What criticism of the Kramer et al article will almost certainly do is decrease the scientific community’s access to, and interaction with, one of the largest and richest sources of data on human behavior in existence. You can certainly take a dim view of Facebook as a company if you like, and you’re free to critique the way they do business to your heart’s content. But haranguing Facebook and other companies like it for publicly disclosing scientifically interesting results of experiments that it is already constantly conducting anyway–and that are directly responsible for many of the positive aspects of the user experience–is not likely to accomplish anything useful. If anything, it’ll only ensure that, going forward, all of Facebook’s societally relevant experimental research is done in the dark, where nobody outside the company can ever find out–or complain–about it.

[UPDATE July 1st: I’ve posted some additional thoughts in a second post here.]

not really a pyramid scheme; maybe a giant cesspool of little white lies?

There’s a long tradition in the academic blogosphere (and the offlinesphere too, I presume) of complaining that academia is a pyramid scheme. In a strict sense, I guess you could liken academia to a pyramid scheme, inasmuch as there are fewer open positions at each ascending level, and supply generally exceeds demand. But as The Prodigal Academic points out in a post today, this phenomenon is hardly exclusive to academia:

I guess I don’t really see much difference between academic job hunting, and job hunting in general. Starting out with undergrad admissions, there are many more qualified people for desirable positions than available slots. Who gets those slots is a matter of hard work (to get qualified) and luck (to be one of the qualified people who is “chosen”). So how is the TT any different from grad school admissions (in ANY prestige program), law firm partnership, company CEO, professional artist/athlete/performer, attending physician, investment banking, etc? The pool of qualified applicants is many times larger than the number of slots, and there are desirable perks to success (money/prestige/fame/security/intellectual freedom) making the supply of those willing to try for the goal pretty much infinite.

Maybe I have rose colored glasses on because I have always been lucky enough to find a position in research, but there are no guarantees in life. When I was interviewing in industry, I saw many really interesting jobs available to science PhD holders that were not in research. If I hadn’t gone to National Lab, I would have been happy to take on one of those instead. Sure, my life would be different, but it wouldn’t make my PhD a waste of time or a failed opportunity.

For the most part, I agree with this sentiment. I love doing research, and can’t imagine ever voluntarily leaving academia. But If I do end up having to leave–meaning, if I can’t find a faculty position when I go on the job market in the next year or two–I don’t think it’ll be the end of the world. I see job ads in industry all the time that looks really interesting, and on some level, I think I’d find almost any job that involves creative analysis of very large datasets (which there are plenty of these days!) pretty gratifying. And no matter what happens, I don’t think I’d ever view the time I’ve spent on my PhD and postdoc training as a waste of time, for the simple reason that I’ve really enjoyed most of it (there are, of course, the nasty bits, like writing the Nth chapter of a dissertation–but those are transient, fortunately). So in that sense, I think all the talk about academia being a pyramid scheme is kind of silly.

That said, there is one sticking point to the standard pyramid scheme argument I do agree with, which is that, when you’re starting out as a graduate student, no one really goes out of their way to tell you what the odds of getting a tenure-track faculty position actually are (and they’re not good). The problem being that most of the professors that prospective graduate students have interacted with, either as undergraduates, or in the context of applying to grad school, are precisely those lucky souls who’ve managed to secure faculty positions. So the difficulty of obtaining the same type of position isn’t always very salient to them.

I’m not saying faculty members lie outright to prospective graduate students, of course; I don’t doubt that if you asked most faculty point blank “what proportion of students in your department have managed to find tenure-track positions,” they’d give you an honest answer. But when you’re 22 or 23 years old (and yes, I recognize some graduate students are much older, but this is the mode) and you’re thinking of a career in research, it doesn’t always occur to you to ask that question. And naturally, departments that are trying to recruit your services are unlikely to begin their pitch by saying, “in the past 10 years, only about 12% of our graduates have gone on to tenure-track faculty positions”. So in that sense, I don’t think new graduate students are always aware of just how difficult it is to obtain an independent research position, statistically speaking. That’s not a problem for the (many) graduate students who don’t really have any intention of going into academia anyway, but I do think a large part of the disillusionment graduate students often experience is about the realization that you can bust your ass for five or six years working sixty hours a week, and still have no guarantee of finding a research job when you’re done. And that could be avoided to some extent by making a concerted effort to inform students up front of the odds they face if they’re planning on going down that path. So long as that information is made readily available, I don’t really see a problem.

Having said that, I’m now going to blatantly contradict myself (so what if I do? I am large! I contain multitudes!). You could, I think, reasonably argue that this type of deception isn’t really a problem, and that it’s actually necessary. For one thing, the white lies cut both ways. It isn’t just faculty who conveniently forget to mention that relatively few students will successfully obtain tenure-track positions; many graduate students nod and smile when asked if they’re planning a career in research, despite having no intention of continuing down that path past the PhD. I’ve occasionally heard faculty members complain that they need to do a better job filtering out those applicants who really truly are interested in a career in research, because they’re losing a lot of students to industry at the tail end. But I think this kind of magical mind-reading filter is a pipe dream, for precisely the reasons outlined above: if faculty aren’t willing to begin their recruitment speeches by saying “most of you probably won’t get research positions even if you want them,” they shouldn’t really complain when most students don’t come right out and say “actually, I just want a PhD because I think it’ll be something interesting to do for a few years and then I’ll be able to find a decent job with better hours later”.

The reality is that the whole enterprise may actually require subtle misdirection about people’s intentions. If every student applying to grad school knew exactly what the odds of getting a research position were, I imagine many fewer people who were serious about research would bother applying; you’d then get predominantly people who don’t really want to do research anyway. And if you could magically weed out the students who don’t want to do research, then (a) there probably wouldn’t be enough highly qualified students left to keep research programs afloat, and/or (b) there would be even more candidates applying for research positions, making things even harder for those students who do want careers in research. There’s probably no magical allocation of resources that optimizes everyone’s needs simultaneously; it could be that we’re more or less at a stable equilibrium point built on little white lies.

tl;dr : I don’t think academia is really a pyramid scheme; more like a giant cesspool of little white lies and subtle misinformation that indirectly serves most people’s interests. So, basically, it’s kind of like most other domains of life that involve interactions between many groups of people.

fMRI, not coming to a courtroom near you so soon after all

That’s a terribly constructed title, I know, but bear with me. A couple of weeks ago I blogged about a courtroom case in Tennessee where the defense was trying to introduce fMRI to the courtroom as a way of proving the defendant’s innocence (his brain, apparently, showed no signs of guilt). The judge’s verdict is now in, and…. fMRI is out. In United States v. Lorne Semrau, Judge Pham recommended that the government’s motion to exclude fMRI scans from consideration be granted. That’s the outcome I think most respectable cognitive neuroscientists were hoping for; as many people associated with the case or interviewed about it have noted (and as the judge recognized), there just isn’t a shred of evidence to suggest that fMRI has any utility as a lie detector in real-world situations.

The judge’s decision, which you can download in PDF form here (hat-tip: Thomas Nadelhoffer), is really quite elegant, and worth reading (or at least skimming through). He even manages some subtle snark in places. For instance (my italics):

Regarding the existence and maintenance of standards, Dr. Laken testified as to the protocols and controlling standards that he uses for his own exams. Because the use of fMRI-based lie detection is still in its early stages of development, standards controlling the real-life application have not yet been established. Without such standards, a court cannot adequately evaluate the reliability of a particular lie detection examination. Cordoba, 194 F.3d at 1061. Assuming, arguendo, that the standards testified to by Dr. Laken could satisfy Daubert, it appears that Dr. Laken violated his own protocols when he re-scanned Dr. Semrau on the AIMS tests SIQs, after Dr. Semrau was found “deceptive” on the first AIMS tests scan. None of the studies cited by Dr. Laken involved the subject taking a second exam after being found to have been deceptive on the first exam. His decision to conduct a third test begs the question whether a fourth scan would have revealed Dr. Semrau to be deceptive again.

The absence of real-life error rates, lack of controlling standards in the industry for real-life exams, and Dr. Laken’s apparent deviation from his own protocols are negative factors in the analysis of whether fMRI-based lie detection is scientifically valid. See Bonds, 12 F.3d at 560.

The reference here is to the fact that Laken and his company scanned Semrau (the defendant) on three separate occasions. The first two scans were planned ahead of time, but the third apparently wasn’t:

From the first scan, which included SIQs relating to defrauding the government, the results showed that Dr. Semrau was “not deceptive.” However, from the second scan, which included SIQs relating to AIMS tests, the results showed that Dr. Semrau was “being deceptive.” According to Dr. Laken, “testing indicates that a positive test result in a person purporting to tell the truth is accurate only 6% of the time.” Dr. Laken also believed that the second scan may have been affected by Dr. Semrau’s fatigue. Based on his findings on the second test, Dr. Laken suggested that Dr. Semrau be administered another fMRI test on the AIMS tests topic, but this time with shorter questions and conducted later in the day to reduce the effects of fatigue. … The third scan was conducted on January 12, 2010 at around 7:00 p.m., and according to Dr. Laken, Dr. Semrau tolerated it well and did not express any fatigue. Dr. Laken reviewed this data on January 18, 2010, and concluded that Dr. Semrau was not deceptive. He further stated that based on his prior studies, “a finding such as this is 100% accurate in determining truthfulness from a truthful person.”

I may very well be misunderstanding something here (and so might the judge), but if the positive predictive value of the test is only 6%, I’m guessing that the probability that the test is seriously miscalibrated is somewhat higher than 6%. Especially since the base rate for lying among people who are accused of committing serious fraud is probably reasonably high (this matters, because when base rates are very low, low positive predictive values are not unexpected). But then, no one really knows how to calibrate these tests properly, because the data you’d need to do that simply don’t exist. Serious validation of fMRI as a tool for lie detection would require assembling a large set of brain scans from defendants accused of various crimes (real crimes, not simulated ones) and using that data to predict whether those defendants were ultimately found guilty or not. There really isn’t any substitute for doing a serious study of that sort, but as far as I know, no one’s done it yet. Fortunately, the few judges who’ve had to rule on the courtroom use of fMRI seem to recognize that.

Regarding the existence and maintenance of standards, Dr. Laken testified as to the protocols and controlling standards that he uses for his own exams. Because the use of fMRI-based lie detection is still in its early stages of development, standards controlling the real-life application have not yet been established. Without such standards, a court cannot adequately evaluate the reliability of a particular lie detection examination. Cordoba, 194 F.3d at 1061. Assuming, arguendo, that the standards testified to by Dr. Laken could satisfy Daubert, it appears that Dr. Laken violated his own protocols when he re-scanned Dr. Semrau on the AIMS tests SIQs, after Dr. Semrau was found “deceptive” on the first AIMS tests scan. None of the studies cited by Dr. Laken involved the subject taking a second exam after being found to have been deceptive on the first exam. His decision to conduct a third test begs the question whether a fourth scan would have revealed Dr. Semrau to be deceptive again.
The absence of real-life error rates, lack of controlling standards in the industry for real-life exams, and Dr. Laken’s apparent deviation from his own protocols are negative factors in the analysis of whether fMRI-based lie detection is scientifically valid. See Bonds, 12 F.3d at 560