It’s Not All About the Facts…

By Jen Nguyen

In many ways, I am living a dream. I’m content. I don’t at all mean that I wish to remain a grad student forever. And I realize that my day-to-day is imperfect. But ask me how I’m doing, and I would tell you I’m “great!” and truly mean it.

There was a point though, when I was pretty down. I had started a new project and felt way out of my element. I couldn’t see where it was going. And as I tried to push through it, I only sank deeper into this sense of detachment. I felt myself changing, but in ways I didn’t desire. (It didn’t help that it freakishly cold outside.)

Thankfully, around that time, I encountered a series of reminders about what I love about science and what I want from my PhD.

One came in the form of the most powerfully affecting talk I have ever experienced. In presenting her work, the speaker reminded me of the fascinating miracle that is life, and why I’d chosen to commit a great deal of my own to study a very small slice of it. Yet, while her work was of considerable merit and her scientific interests aligned nicely with mine, it was how she spoke of them that struck me. I found myself motivated, shaking with energy, brimming with tears.

How did she do this? Never would I have expected to feel such emotion in response to data …or I should say, someone else’s data. (I’ve certainly felt surges of adrenaline when looking through my own.)

It’s not that my chosen questions are of greater “importance,” scientific interest, or social impact than (most) others. (Please afford me a few biases.) I also don’t believe that I have a greater personal investment in my work than others have in theirs. The difference, I think, stems from the widespread notion that science is purely objective.

Of course, there is great precedent for objectivity. I too admire the elegance of those who win us over, not with flourish, but with reason. During this talk however, logic, clarity, and rigor (while exceptional) were not what caused me to well up inside. The trigger, again, was her description of her subject matter. To her, it wasn’t “matter”. It was alive. It fascinated her, and she spoke as if unable to suppress her regard for it.

Kathryn’s seminar reminded me that I’m after not just results, but an experience. In a few years, I don’t want to measure the significance of my PhD by what I learn about bacterial growth (though it’s pretty damn incredible) or by who I can persuade into caring about it. Instead, I want to recall how I did it and why I loved it. With these, I know I will leave MIT more capable and confident than when I started. To uncover some earth shattering science, or to convince the world of the power of microbes – well, that’s the icing on the cake.

Artwork: David Goodsell

The science of disbelief

By Sarah Spencer

In late January, the Pew Research Center conducted a survey of the American public regarding their views on science and society.  The results showed strong positive support of the scientific endeavor, with 79% of adults claiming that science has made life easier for most people, and 72% stating that government investment in research pays off in the long run.  So why the pushback?  Why is government funding stagnating while huge public campaigns propagandize against scientific results?

Of course these issues are complicated, but a core driving force involves the psychology of how we maintain and reinforce our beliefs.  And the truth is, we maintain and reinforce our beliefs (yes, even scientists).  Pre-existing belief structures have a strong influence on the facts we consider important, the methods we employ, and what we conclude from data.  This ‘motivated reasoning’ underlies the difficulty of arguing about ideas with facts.  A fact, or even a paper, cannot easily overturn a lifetime of built-up belief structures.

There is support for belief-driven fact filtering from many, many social psychology studies over the past 40 years.  A foundational 1979 paper gathered a group of individuals with strong beliefs either for or against capital punishment.  The researchers exposed the group to fake research articles either supporting or discrediting the ability of the death penalty to reduce crime.  As expected, individuals thought the article supporting their beliefs was more convincing, and strongly criticized the article which refuted their beliefs.  Since this publication, researchers demonstrated belief-directed reasoning in just about any issue you can think of: affirmative action, gun rights, weapons of mass destruction, etc.

A common argument against rampant ‘motivated reasoning’ points to an individual’s level of scientific training.  Can people without deep scientific training truly quality-filter scientific evidence and avoid being misled?  A recent Nature Climate Change letter discredits this idea using climate change as an example.  Surprisingly, members of the public with the greatest scientific literacy and technical reasoning became the most polarized on the issue.  More training resulted in stronger opinions on both sides of the argument.  These knowledgeable individuals simply used the facts to support their pre-conceived beliefs.

Does this mean we’re all inherently illogical?  Of course not – people still accept new evidence and change their minds every day.  It just means we have a multitude of priorities to maintain.  There are people who care about accuracy, supporting a conclusion, maintaining connection with their friends, and/or bolstering their own sense of identity.  Any of these priorities can, at some point, interfere with perfectly logical interpretation of fact.  Combine personal priorities with this whole underworld of instantaneous emotional response to contrary evidence, and it’s understandable why we have a hard time always reaching the same conclusion.

Thankfully there are ways to get the facts straight, but they sometimes rely on setting facts aside.  We live in a world where facts do not reliably persuade, so one useful alternative is to appeal to values.  As an example, if you present a group with the same article under two titles:

“Scientific Panel Recommends Anti-Pollution Solution to Global Warming”

“Scientific Panel Recommends Nuclear Solution to Global Warming”

the latter title results in more people convinced that humans are causing global warming (Kahan et al., 2007).  Something about the focus on industry appeals to the emotions and priorities of climate change disbelievers.  Sure, things are complicated and messy and hard to measure: it’s human psychology.  Still, the next time you’re arguing for a controversial idea, consider structuring around values and then adding data.  This might avoid immediate denial and get more people appreciating the scientific endeavors they implicitly support.

 

Should I do a postdoc?

By Felix Moser

In this post, I examine the “postdoc” as a step in a young scientist’s career. A “postdoc”, or one’s time as a Postdoctoral Associate in a lab, is traditionally the phase of a scientist’s career immediately after graduate school that precedes a career in academia. But due to sparse availability of faculty positions and plentiful PhDs produced by graduate schools every year, a career in academia is no longer a realistic option for the vast majority of new PhDs. In the face of this fact, does doing a postdoc make sense? When is it valuable for a new PhD to pursue a postdoc?

There are essentially four reasons to do a postdoc:
1) Because you want to.
2) To get additional training (even though you may not want or need it).
3) To prepare for the next step.
4) Your PI kicked you out and the coffee shop next door ain’t hiring.

The first reason is perhaps the most obvious and least acknowledged. Many people take postdoctoral positions simply because they like the features of such a position. Because we scientists and engineers are driven primarily by our curiosity, we tend to want to learn new things. So, getting additional training outside of our graduate field is exciting to us. Some people relish to opportunity to hone their skillset. Others get excited about a particular project, lab, or location. Some people simply love the freedom of the academic environment, and, barring available faculty or scientist positions, find the postdoc as the easiest way to extend their time in academia. Looking only at these features, a postdoc may seem like a great job. Unfortunately, with a pay of ~$40k/year, the risk of a bad PI, poor funding, or bad project, and slim chance at career advancement in academia, these features come at some cost.

The second reason to do a postdoc, additional training, is typically the official one given by academic institutions and companies who won’t hire people straight out of grad school. Even the lucky few who get hired to faculty straight out of grad school are often asked to postdoc in another lab for a year to round out their experience. Part of the reasoning for the additional training is that science is so much bigger and complex now than it was ~30 years ago that it simply requires more experience in different subjects to really produce someone that can lead a research program. The unofficial reasoning is market forces. In a market where there is no shortage of excellent people WITH postdoc experience, why hire someone without it?

The advantages to additional training are obvious. Developing skills and knowledge complementary to your current skill/knowledge set should add to your value as a scientist. You’ll be more likely to build bridges between fields and carry out true innovation. You should become a stronger analyst and problem-solver by learning new systems, different methods, and new lines of thought. However, some drawbacks should be acknowledged. An old adage about young scientists maintains that “they don’t know what can’t be done”. Someone who has been around the block a few times might know the neighborhood very well, but they might be less inclined to head into a new part of town. Also, not all training is equally effective or useful.

The third reason to do a postdoc is to prepare for the next step in your career. The postdoc should position you for the job you want after. Because of this, you should tailor your postdoc to land the job you want. Regardless of whether you are targetting an academic or industry positions, plentiful innovative research, publications, and excellent communication are a must.

If you want to become a professor, a successful postdoctoral period is a cornerstone of your application. Unless you’ve had an exceptional graduate career, don’t bother applying for faculty without at least one postdoc under your belt. The postdoc is in part considered a time in which you are thought to be skilled enough to be fairly independent. Hence, it is a type of trial period for you as an independent researcher and a glimpse at what you would be like as a group leader.

BE faculty have told me that the most important thing for a postdoc is to “make your project your own.” This speaks to the independence and ability that’s expected of an academic researcher to develop their own science and put together a compelling story. A really successful postdoc will take a project idea and develop it in a way that is characteristically their own, so when people read your published work, they will link its ideas to you. A successful postdoc will set your faculty application apart from the 200+ others an institution will receive. It will do so foremost in 1) ability to publish excellent science, 2) creative, thought-out ideas, and 3) the ability to tell a compelling story (translates into good grantsmanship). Even with these, an excellent rec letter is required to seal the deal. The rec letter, like any networking connection, should be from a recognized and trusted PI. If this comes across as a sort of “good ol’ boy” network, that’s because it is. When there are 200 CV’s on the table, 10% of which excel in #’s 1-3, the one that’s picked will likely be on someone’s recommendation. Because of this, an academic postdoc is best done for a PI who is well-respected in the field in which you intend to become faculty.

To tailor your postdoc for an industrial positions, there are some things to keep in mind. For one thing, the rec letter doesn’t matter nearly as much as it does when you’re applying for jobs in academia. Industry mostly cares that you can do your job well and less so about what your PI thinks of you. Also, the PI may not be well known in industry, despite stardom in academia. You should also be mindful that some industry positions require specific skills which, if you lack them, you should seek to acquire during the postdoc.

The length of a postdoc that prepares you for an industry job is another key difference. Depending on your goals for an industry job, the postdoc can be more flexible in length. If, for example, you are looking to postdoc solely to learn a skill-set, then you should call the postdoc complete as soon as that skillset is learned and a job is secured. If your goal is ANY job, then the postdoc is merely a stepping stone. Once that first job is locked down, you’re done. The postdoctoral advisor should be kept in the loop at all times, however. You don’t want to burn a bridge with a connection that could be a great help to you later.

Doing a postdoc for preparation for an industrial job may lead to some friction with the academic advisor. Old-timey measures of success value placing students and postdocs in academic positions. An industry position is considered a failure. This is an antiquated attitude and any advisors that retain or propagate it should be avoided, if not confronted. Their perceived interests are completely misaligned with that of the postdoc seeking an industry position, and the postdoc should not expect support from them. Advisors that are supportive of your goals, whatever they are, should be sought after. However, it is up to the postdoc to be as clear with the advisor as possible about their goals. The key is to be proactive and open in communicating with the advisor.

Industrial postdoc positions should be approached with special caution. If you are angling for a faculty position, you want to make sure an industrial postdoc will allow you to publish. Be aware that if and what you publish is often strictly controlled. The company IP lawyers may very well veto any publication that divulges sensitive information. If you are angling for a permanent position in the company, you should get explicit confirmation that similar positions have in the past led to permanent hires. Try to find and talk to people who previously completed postdocs with the company. The worst industrial postdoc programs (may also be called “internships”) are simply ways to get a scientist on the cheap. These may string you along for years without any promotion and offer little real security. Also be wary of signing a “non-compete” agreement. These agreements will bind you legally from working in your field outside of the company. Because scientists are typically highly specialized, resigning your right to employ your skill set outside of one company can be deadly to your career (unless you move to CA, which doesn’t enforce non-competes).

Finally, it should be acknowledged that a big reason many people do a postdoc is because they simply can’t find any other work. Today’s market for scientists is more competitive than it has ever been, both in academia and industry. The market is flooded constantly with many more PhD’s than there are available positions. Because the NIH and other funding agencies set the floor for postdoc salaries well below industry standards, it is much, much easier to find a postdoc position (~$40k/year) than a full-time scientist position (~$70-80k/year starting). If you remove the postdoc from the table of possible next steps for graduating PhDs, then becoming a barista starts to look like a plausible alternative.

PCR of the heart, a REFS story

By Scott Olesen

If you’re the kind of scientist who does a lot of PCR, a common experimental method for making a few strings of DNA into many strings, you’re probably worried about chimeras. PCR makes many faithful copies of the original DNA strings, but is also makes a few chimeras, combinations of two or more of the original strings. It’s like you tried to clone a human and a horse in the same vat and ended up with many human clones, many horse clones, and a few centaurs.

One way to remove chimeras is to look at each string of DNA in the PCR product and ask, “Is this string just a combination of two other (more plentiful) strings?” Thankfully, only a few strings in the product are chimeras: it’s easy to notice that a centaur is half human, half horse when there are many more humans and horses than centaurs.

It’s tempting to believe that we are dispassionate observers of facts, that we perceive humans and horses but not centaurs. After I took conflict management training as part of the REFS program, a confidential conflict coaching resource for BE grads, I knew the opposite was true. When it comes to interpersonal matters, the facts—who said what, when, with what tone and what facial expression and in what context—get all jumbled.  I remember a few of the bare, dry facts—the humans and horses—but most of what I have in my head are centaurs and reverse-centaurs and other monstrous things that look nothing like the humans and horses they supposedly come from.

Take me and my parents, for instance. When I was a teenager and my parents fought, I would protect myself and punish my parents by withdrawing into sullen, spiteful silence. I’ve tried to give up that tactic, and I’d thought I’d succeeded, but from time to time through the years my parents have said and done things that made me sure they thought I still tried to punish them with my sullenness. I experimented with a number of schemes to un-convince them—like cheerily breezing past their connubial spats or, when the mood was bright, talking in the abstract about “how much I’d changed” since my teenage years—none of which involve actually talking about the behavior in question.

With DNA, you just look at each string of DNA and check if it’s a combination of other strings. REFS taught me how straightforward, effective, and incredibly terrifying it can be to just ask, “Hey, are you a centaur?” During a car ride during Christmas break, my parents had a little fight with each other and I activated my defense mechanism, sullen silence. In that moment, I thought “Oh no, they’re going to think I’m trying to punish them!” I decided I hadn’t had 40 hours of conflict training for nothing, so I took a deep, emotionally-laden breath and blurted, “I’m worried that, when I’m quiet, you think I’m being intentionally sullen.”

My parents were surprised. “Oh, we don’t think that! I mean, you definitely were when you were a teenager, but not any more.” Poof. Adios, centauro.

My point is that what the Refs have to say about conflict—basically, to get in there and say how you feel—will probably sound really obvious. The hard part is doing what’s obvious. In science, it’s often easy to do the project if you know the best way to do it, but it takes a year to troubleshoot the protocol. In life, I think the best way is often the most obvious and the most difficult.