I’ve been watching an interesting discussion unfolding at DrugMonkey, prompted by a post at Science Bear’s Cave, about whether not irritating your lab group’s principal investigator ought to be your highest priority. As DrugMonkey notes, such a strategy can have bad consequences:
If there is a scientific trainee who fears to mention to the Boss that the printers aren’t working, this trainee sure as hell isn’t going to mention “Oh gee, I think that figure you are so amped about from that other postdoc is totally faked”. And who knows how far this PI-pleasing attitude might carry one.
Is the desire to keep the boss PI happy greater than any affection for, say, genuine data?
The discussion in the comments covered some familiar ground about the general unpleasantness of the experience of science as gladiatorial combat, the relative powerlessness of the scientific trainee relative to the PI or even the most favored postdoc, the need to believe in your own results (even if, objectively, you should not) in a climate of shrinking funding, whether science is or is not a tea party peopled with Care Bears (and if so, what exactly that would entail) …
For some reason, reading this exchange, I slipped into my committee-chair mindset. (Of course, this is very odd, seeing as how my current sabbatical has included a break from committee service. Could it be that I miss it?) I thought to myself, “The people here have rational reasons for the ways they’re behaving, even if those behaviors lead to predictably bad results. Are there any easy-to-implement changes that would make it rational for them to behave in ways that probably bring about better results?”
Then, I slipped into pedagogy mode (since I’m also on a hiatus from teaching) and thought, “Students focus on learning the things you test them on, not the things you say are important for them to learn. And when you don’t give them any sensible information about what you’re evaluating, or about the basis for evaluation, they will become bundles of hate, fear, and frustration.”
At this point, I had a flash of insight to a change that I think has potential. I can’t guarantee that it would work, but I think it would be foolish not to try it.
What if scientific mentors (who are engaged in the dual projects of producing new scientific knowledge and producing new scientists) routinely articulated their expectations of the scientific work and of the trainees participating in that work? What if being a trainee didn’t feel like an n-year examination (written, oral, and lab-practical) for which the grading criteria were a closely-guarded secret?
Would this be more likely to create lab groups in which people shared information about what’s gone wrong with experiments, or about problems working up the data, or about crucial information from the literature found later than one would like?
Would it foster an environment in which people were less likely to try to bluff their way through interactions, more likely to ask for guidance when they needed it?
Could it create conditions in which the task of becoming a grown-up scientist was understood as a shared project in which trainee and mentor each need to shoulder part of the burden?
Might it make more obvious the separation between the two projects — for example, the possibility that one’s skills as a scientist can grow even as one’s attempts to get a particular research project to work are thwarted?
Undoubtedly, there are PIs out there now who are reasonably open about their expectations and fairly hands-on in interactions with trainees around their shared scientific inquiries. My sense, though, is that they are rare enough that students may find themselves asking, in the presence of such mentors, “He says he wants X, Y, and Z from me, but what does he really want? Will X, Y, and Z be enough, or is there some extra thing that I’m supposed to be smart enough to figure out that he wants, too?”
If the normal state of affairs were to lay all cards on the table (at least within research groups and training relationships), would members of research groups be able to spend more time focused on the project of building scientific knowledge and less time playing against each other?
If so, how do we shift the normal state of affairs from where it seems to be at the moment?
What if scientific mentors (who are engaged in the dual projects of producing new scientific knowledge and producing new scientists) routinely articulated their expectations of the scientific work and of the trainees participating in that work? What if being a trainee didn’t feel like an n-year examination (written, oral, and lab-practical) for which the grading criteria were a closely-guarded secret?
I disagree that keeping these standards a secret is the norm, rather than the exception.
Great post Janet!
What if scientific mentors (who are engaged in the dual projects of producing new scientific knowledge and producing new scientists) routinely articulated their expectations of the scientific work and of the trainees participating in that work? What if being a trainee didn’t feel like an n-year examination (written, oral, and lab-practical) for which the grading criteria were a closely-guarded secret?
Actually, I remember my advisor saying almost precisely this after my candidacy exam…the exam itself went really well but my advisor was frustrated with the fact that it felt very adversarial, that I was always having to defend my ideas rather than having a more collaborative interaction with the exam committee. I would love to lean towards cooperative rather than adversarial interactions…but in my personally anecdotal experience it rarely happens this way.
CPP – I would love to work for you if you think that secret standards are the exception. That says a lot about how YOU feel you are communicating with your people. Do they feel the same way?
In my experience, my advisor feels that he communicates quite clearly and makes his expectations known (I know this because he has said so explicitly), and yet I frequently feel as if I’m being blind-sided by criticism to the effect that I have failed to meet some mysterious expectations…and most of our interactions feel (to me) like I’m being grilled rather than mentored. I wouldn’t say that it’s openly hostile, but it certainly not the most effective and efficient way to conduct a professional relationship, or research in general.
So much of it has to do with the relative empowerment (or lack thereof) of both parties and much of it has to do with the fact that people with disparate communication styles have a lot of difficulty implementing mutually satisfactory interactions.
I wish that I had some idea of how to implement some kind of sea change. I don’t. For now, all I can say is that I see the problem and I hope that I will still be aware of it as I progress into a leadership role and conduct my interactions in a way that alleviates this kind of tension.
I can say that I have worked for people that treat their “underlings” like members of a team rather than underlings and this makes a tremendous difference as compared to those who tend to hand down directions and judgments from on high. I think that these are primarily differences in individual leadership and communication skills…how we can push for more individuals with effective skills in these areas I don’t know. Perhaps request letters of reference in hiring searches from people who have worked under the candidates as well as the colleagues? And then value these LORs as highly as those from people in power? I think it would be great but I doubt it would actually happen – still very pie in the sky. What matters most is publication and funding records because this translates to revenue for the institution. Until there is a concrete and tangible reward system in place that specifically values and rewards good mentoring relationships, I think that it will continue to be overlooked as a criteria for hiring and promotion. This will not affect the ones who do a good job of it already, but as long as there is no incentive for change for those who don’t, there will remain those who do a poor job of it.
To be honest, I’m having trouble understanding what possible kinds of “secret” or “mysterious” expectations there could possibly be. Trainees need to (1) learn the relevant literature, (2) learn to be experimentally productive, (3) learn to give decent presentations of their work, (4) learn to write decently (grants and manuscripts), (5) learn to design experiments, (6) learn to be creative and originate new ideas, (7) learn to effectively mentor less-experienced trainees, (8) learn to communicate effectively on an informal basis with other trainees in the lab and with the PI.
Different trainees exhibit different capacities for mastering these different aspects of becoming a scientist, and I’m sure there are some things I have missed. But I am having a really difficult time thinking of what “secret” or “mysterious” expectations there could be.
But I am having a really difficult time thinking of what “secret” or “mysterious” expectations there could be.
Me too!
All the ones you mention make good sense and are pretty obvious even if not explicitly stated. No problems there. But then even when you’re on top of those (and asking for feedback to be sure you’ve not dropped the ball) and it is still clear that you have disappointed teh mentorz, for reasons unspecified, it seems the only logical conclusion is that you were expected to do…something else?
Also, some confusion can come into play when the trainee has certain standards to define “decent” with respect to presentations, lit, etc. that differ from what the mentor considers “decent”. Sometimes it’s because the trainee is doing it wrong and sometimes it’s just a difference in personal preference. In either case, it helps if whoever holds the expectations can state specifically what is the “right” way or their personal preference, instead of waiting for the trainee to guess.
CPP it is not only about the big-picture expectations, but also about daily, details kind of things. Should a postgrad/postdoc guess the details of experiments that his PI wants? I see it in my current lab, where students are not given much support, or training, with their experiments, and later their results are torn to pieces in the formal presentations, where they find out that they did this and that wrong. But there was no supervision, no feedback during the work was carried out, no opportunity to ask questions, no clear expectations how the experiments were supposed to be done in the first place.
Well, I am now in a very dysfunctional lab, so maybe that is a bad example.
But there are those details that you need to find out from somewhere, for example if there are annual reports for postgrads in your group, how long should they be, 5 pages or 50 pages. If the PI does not tell their students what they expect, how can they later be angry at the students who handed in the shorter reports.
It takes that the PI knows what they expect, and communicate it clearly, and not for their underlings to find out post factum that they did not meet the expectations that they did not know that were there.
It is a very poor method of training to have people mind-read the expectations of their PI.
I agree DrL. CPP outlined the major objectives very well, but I think what is hardest for students/mentees/advisees is when the details are not more explicitly stated. To take even one of the more explicit goals, (1) learn the relevant literature, how does the PI train her/his advisee if s/he does not have literature research skills (I have encountered this in undergrads I mentor)? And if the student presents on the material to demonstrate said ability, how does the PI follow up with or comment on the performance?
To continue with #1, the way I break this down for undergrads is that I have them bring me five articles from the literature on a given topic. We discuss them, then we discuss what a relevant research question would be. Then they go back to the literature for five more articles with a clearer focus. We meet again, and then I start holding their hands less. They come up with another ten articles, then they write an annotated bib, then they present their findings to me, then they write it up like a lab report.
Grads on the other hand don’t have the first few steps done with them because I assume they know how to find articles. We discuss the topic together and then they go straight to the annotated bib. If we need to backtrack to help them better design a research question we do that. And so on.
What I’m trying to demonstrate is that in order to train students in the objectives CPP articulates we need to do a lot more planning. What I suspect happens, in those labs where rules or objectives appear to be secret or unknowable, is that the PI has not given sufficient thought towards HOW one achieves these goals. So then when a student presents poorly the PI rips him/her to shreds because s/he is angry that the student failed. I suppose another explanation could be that that’s the model that PI was trained in, because learning how to mentor is not exactly part of the tenure review package. But I know PIs who were trained in warm fuzzy labs and then do the last-minute mentoring/tearing people to shreds thing.
So as not to hijack comments I’ve decided to write my own post in response to this one. You can find it here: http://k8grrl.blogspot.com/2009/02/what-does-good-mentorship-look-like.html.
Pingity Free-Ride!