18 July 2017

Rough rides at tenure time


Yesterday, I wrote about Dr. Becca’s tumultuous ride through her tenure process. (And thank you to all who read, like, tweeted, shared, and commented!) Becca has done many early career scientists a favour by documenting this difficult process.

I’ve talked from time to time about how important it is to share our failures. But we particularly don’t like to draw attention to issues that came up at tenure time. I wrote about my problems with tenure after I squeaked through the process. I was not writing a pseudonym, and I didn’t blog about the process much while I was going through it. I was very mad about it then. I don’t get visibly upset talking about like I used to, but I can’t say I’ve made peace with it. With the better part of a decade between then and now, I can see why I got a hard time at tenure, but I still feel I was not treated well.

Terry McGlynn had an even rougher time. He was denied tenure, which he wrote about extensively in the Chronicle of Higher Education.

But what I haven’t done specifically is to talk about what came afterwards for me. A lot of people think that after academics get tenure, they drop off and take it easy.

I got better.

After tenure, I finally had everything in place. The gears were turning, and I started to get the researcher coming out much more consistently, with more original data driven papers. And I had seen the adage, “Don’t let the perfect be the enemy of the good.” I lowered my standards and stopped waiting for projects to get that one last bit of data. And stuff started to happen for me. I became one of the most published faculty in the department.

I am not trying to brag here. I know many people would look at my research track record and deem it second rate (at best). “Sand crabs, Zen? Nobody cares about your sand crabs.”

The moral of the story? It’s to remind people that trouble, even at this critical point in an academic career, does not have to cripple the rest of your career.

And that publishing well is the best revenge.

Related posts

Now part of the problem
Low points
Nevertheless, she persisted

External links

Coming out of the closet, tenure denial edition

17 July 2017

Nevertheless, she persisted

Sometimes, you get to watch a friend win one. And that win is practically as sweet as one of your own.

Friend of the blog Dr. Becca has been getting a rough ride at tenure time. Until today:

THE PROVOST REVERSED HIS DECISION AND IS RECOMMENDING ME FOR TENURE

First things first: Congratulations, Becca! I am so happy for you! Wooo!

Other things: Becca’s win is important beyond just the obvious significance for her and her students and collaborators. It needs to be seen and discussed widely for two reasons.

First, her case needs to be talked about because the grief she was getting was all about one thing: money. Scratch that: it was because she didn’t get the right kind of money. Her job was being threatened because she hadn’t brought in a stand alone research grant from the National Institutes of Health (an NIH R01, to use the jargon).

Becca’s situation is the nightmare scenario that many early career scientists are staring down. The NIH budget is flat, applications are up, and most recognize that the success rate in applying for NIH grants is now so low that many perfectly good projects go unfunded.

In other words, getting a grant has a healthy dose of luck to it and no amount of granting savvy can ensure you will pull down any particular grant. Lack of a grant does not mean your colleagues don’t think you’re doing crummy science.

Becca’s situation shows how dire and destructive this habit of “outsourcing” tenure decisions to granting agencies has become. Professors and administrators need to talk about this and adjust their expectations to line up with reality, and not expect the stone to give blood if you “incentivize” the stone enough.

This is something that has been buzzing in the background for a long time, but the situation has worsened in the last 6-7 years. Academics are used to stability at much longer time scales and aren’t prepare to adjust to the ground shifting underfoot in the time it takes to hire a professor to her tenure review.

Second, Becca’s case matters more generally than her alone because, as Neil Gaiman (channeling G.K. Chesterton) says:

Fairy tales are more than true: not because they tell us that dragons exist, but because they tell us that dragons can be beaten.

Becca shows that you can fight the dragons of university administration, and you can win. And a lot of early career academics need to know that. Because dragons are big and scary and it is easy to give up and concede the battle.

Becca was confronted with career dragons.

Nevertheless, she persisted.

Related posts

The secret life of a banner
The secret life of a banner, part 2

16 July 2017

The future is female

This year has seen something special. There’s been a hunger for new heroes. You can see it in these projects.


Hidden Figures. It challenged the pop culture juggernaut Star Wars at the box office, and got Oscar nominations, too.


Wonder Woman. The biggest hit of the summer, still going strong.

And now... the thirteenth Doctor.


It’s going to be fantastic.

Congratulations, Jodie Whitaker! I look forward to seeing you pilot the TARDIS and fight the monsters!

Added: Reaction to the latter.

13 July 2017

Five years for seven points of data

I was very excited yesterday. I got to add another data point to this graph:


It’s taken me five years to get those seven data points. Five. Years.

It’s not for lack of trying. Each data point depends on me catching a rare event. There’s a limited amount I can do to try to catch those rare events, so this graph is building up slowly. It’s not quite a pitch-drop experiment, but I am seriously wondering if I am ever going to have enough data that I will feel confident about publishing it.

I share this because there are a lot of people fretting about the speed of science these days. People want want fast review, and fast publication. Some are turning to pre-prints for greater speed. But sometimes, try as we might like, some questions force you to take a long, slow slog to get to the answer.

12 July 2017

The bat signal: Can cricket ears hear their predators?

(This was originally published here in 2005.)

Few events in animal behaviour evoke an observer’s visceral response as interactions between predators and prey, leading to poetic metaphors like, “nature red in tooth and claw.” The mechanisms through which prey avoid being caught and eaten provide some of the best examples of behaviours whose neural basis is reasonably well understood. For example, in fish, the Mauthner cells are key players in generating C-start escape responses (Korn and Faber 2005); in crayfish, the lateral and medial giant interneurons generate escape tailflips (Edwards et al. 1999). Surprisingly, however, our knowledge of when these well studied circuits are triggered by actual predators in the wild is rather limited, though those gaps are beginning to close (Herberholz et al. 2004).

Crickets have neurons that trigger escape responses, named AN2 (also referred to as Int-1). Unlike fishes’ Mauther neurons or crayfish’s giant interneurons, which can be triggered by a wide range of sudden stimuli, AN2 neurons appear to serve as detectors for one particular type of predator, namely echolocating bats (Nolen and Hoy 1984, 1987). While AN2 neurons respond to a wide range of sound frequencies, they are particularly sensitive to ultrasound, that is, sound frequencies that are too high for human ears to hear (Nolen and Hoy 1987). This is the approximately the same range of sound frequencies that echolocating bats use when foraging. But, as a recent paper by Fullard and colleagues (Fullard et al. 2005) notes, the key word is “approximately.” There are many species of bats, which differ in their foraging tactics, and emit a wide range of sounds as they do so. Most lab studies, for understandable reasons of simplicity and convenience, have used pure tones generated by computers to trigger crickets’ auditory neurons.

Fullard and colleagues studied Teleogryllus oceanicus, a cricket species found across much of the western Pacific. They recorded the calls of a half-dozen species of bats that share habitat with this cricket, then recorded AN2 neurons as they played back the bat calls at different sound intensities.

The crickets’ AN2 neurons responded to calls from all six bat species, if the sound intensity was 80 decibels sound pressure level (dB SPL) or more, although they did not react equally to all bat search calls.

Simply firing the AN2 neuron, however, does not determine if the cricket can avoid a foraging bat, because a single spike of AN2 is not sufficient to trigger an escape response (Nolen and Hoy 1984). By examining the pattern of firing in more detail, the authors were able to estimate how far away a bat call might trigger an escape response. Only calls by three of the bat species fired AN2 neurons strongly enough to generate escape responses before the bat would be aware of the cricket's echo.

If the AN2 is indeed a “bat detector,” it is reasonable to hypothesize that it has been shaped by natural selection to detect bat species living in the same habitat. All bat calls tested were from species that live in the same regions as T. oceanicus (i.e., sympatric species), but one might reasonably predict that AN2 should be less responsive to calls of bats that do not live in the same regions (i.e., allopatric species). That T. oceanicus has such a wide distribution, however, might mean that its auditory system has remained a bat “generalist.” Another prediction of the “bat detector” hypothesis would be that the bats that AN2 detects best would be those of species that are the most successful predators of crickets. In this case, the bat species Tadarida australis generated the greatest AN2 responses, raising the question of what the natural ecological interactions are between the cricket and the bat.

The bat species that is arguably the least conspicuous to crickets demonstrates the importance of understanding natural ecology in interpreting patterns of neural activity. Of the six species of bat whose calls were tested, the least conspicuous to crickets was Nyctophilus geoffroyi, because the echolocating calls of this species are too short and too high frequency for the crickets’ ears to detect reliably. The simple hypothesis might be that this bat species is a mammalian “stealth bomber:” by using echolocation calls that are almost undetectable by crickets, the bat would seem to be well equipped to pluck crickets from the air at will. Instead, N. geoffroyi seems to forage primarily by “gleaning,” i.e., locating insects by the sounds they emit and picking them off the ground (Bailey and Haythornthwaite 1998), a tactic that circumvents crickets’ tuned AN2 “bat detector” almost entirely.

References

Bailey WJ & Haythornthwaite S. 1998. Risks of calling by the field cricket Teleogryllus oceanicus; potential predation by Australian long-eared bats. 513. Journal of Zoology 244(4): 505-513

Edwards DH, Heitler WJ, & Krasne FB. 1999. Fifty years of a command neuron: the neurobiology of escape behavior in the crayfish. Trends in Neurosciences 22(4): 153-160.


Herberholz J, Sen MM, & Edwards DH. 2004. Escape behavior and escape circuit activation in juvenile crayfish during prey-predator interactions. The Journal of Experimental Biology 207(11): 1855-1863.


Nolen TG & Hoy RR. 1984. Initiation of behavior by single neurons: The role of behavioral context. Science 226(4677): 992-994.

10 July 2017

Goodhart’s Law



When a measure becomes a target, it ceases to be a good measure. - Goodhart’s Law

The longer I’m in academia, the more I appreciate the wisdom of this statement.

07 July 2017

Why I stopped writing grants

A couple of threads on Twitter recently reminded me of something. This from Liang Gao (my emphasis):

Just visited a new PI. He showed me beautiful research, top publications , and thousands pages of unfunded proposals. What the hell is going on?

Then there was Prof-like Substance:

Remember that when applying to NSF, this is what you’re up against. ~6% success from the process. Which is why I tell people over and over and over that if you don't diversify, you will get eaten alive. You. Can. Not. Go up against 6% success for a decade and think everything will be fine. It won’t.

Then there was this charming reminder from Jacquelyn Gill that in addition to dealing with biases about sex, race, and “academic pedigree,” you have to deal with biases about geography:

An equipment grant I'm a co-PI on is #NSFunded! I'm grateful for the chance to do some fun new research. But two reviewers mentioned how small UMaine is. One said it “only has 13,000 students.” Another said there’s “not much up there but moose.”

A lot of people are reaching the point I got to maybe four of five years ago, when I wrote:

Personally, if you’d asked me when I started this job if I thought that I’d be able to get grants for my research, I’d have said, “I think it’ll take me a few tries, but I think I can do it.” Well, that hasn’t happened. So I’ve had to re-invent myself, my expectations, everything, from almost the ground up. It’s been a decade-long battle to redefine myself as a scientist. I’m still not done.
 
I realized that producing thousands of pages of grant proposals was not satisfying for me, either personally or professionally. The odds were long and not improving. People probably think there’s less in South Texas than there is in Maine.

I also realized that managing those grants I did get were not satisfying for me. I’ve complained for a long time that trying to spend a dollar from a grant requires a bottle of aspirin, because it’s an instant headache.

So I mostly quit writing grants. I’m still writing some pre-proposals for NSF, but none have gotten an invitation for a full proposal.

Instead, I have focused on the bit that I find most satisfying for me: writing papers. I have focused on creating “$5 projects” that can go forward, grant or not. My research doesn’t run on money. It runs on willpower.

And I just submitted a manuscript to a journal today, thank you, that was generated with no grant support at all.