One thing that really impressed me was that SDB set aside a whole afternoon for a symposium on education. There was a definite message that we developmental biologists ought to invest more effort in better teaching and public outreach—there is incredibly cool stuff going on in this field and it is going to be fundamental to exciting new fields that are being spawned right now. Unfortunately, the symposium was representative of how we may aspire to get the message out but still fail.

I've been to a lot of these attempts to teach new educational strategies to scientists. They follow a pattern: there's usually some guy who is a professional academic educator with no knowledge of the specific scientific field who gives an enthusiastic talk that uses a barrage of buzzwords. Then there are a series of scientists who give talks on their experiences with Buzzword A or Buzzword B, and sometimes you can glean some useful ideas from them...but in general, it's a painful experience.

Oh, and one thing we always hear: the Lecture is obsolete. There are all these new and powerful didactic tools you can use to connect to your audience, engaging even huge classrooms full of students. The funny thing is that every time, these pronouncements are delivered in a lecture.

This session followed the model exactly.

First up was J Bransford from the University of Washington, not a developmental biologist, but apparently someone with impeccable credentials in education. He gave a very entertaining and energetic lecture on "How People Learn." Alas, I did not learn how people learn. I did get buffeted with phrases I've heard many times before, but haven't found particularly helpful: "knowledge centered learning," anyone? How about "adaptive expertise?" The absolute nadir was a powerpoint slide with a series of bullet-pointed terms, "mosaics", "modules", "granules", and a few others—the information content was precisely 0.

I don't want to be too critical of Bransford, though. While I didn't learn anything from the explicit content of his talk, the meta information was good. As far as I was concerned, he could have been speaking Chinese, but his lecture still kept my attention with good speaking style, a clearly well-prepared and well-laid-out set of slides (with some exceptions), and a lively pattern of expressive changes. No droning bore, he. Although the incessant, florid gestures during the talk were a bit distracting—I was wondering if he was sending secret messages in ASL to the audience.

The best talk of the session was by Bill Wood, on "The Scientific Teacher." His thesis is that we don't do a good job of approaching teaching in the same scientific mode we do our lab work. We stick to the traditional lecture because it's how we learned the material, ignoring the fact that we are actually peculiar outliers, not representative of students as a whole. Alternatives to lectures seem less efficient, which I agree with: as long as our goal is to open up a firehose of information and soak the students with it, nothing beats the good ol' lecture...but maybe that's not what we should be doing. He pointed out that we've institutionalized a process of evaluating teaching that actually hides evidence of problems. Faculty are rewarded for being friendly and entertaining, not for how well they teach. (At my university, we have a standard student evaluation of teaching with five general questions, one of which is to rate the quality of the room—yeah, one (tiny) factor in my tenure review was whether my students thought the room I was assigned to teach in had comfy chairs.) And the big problem is that most educational institutions simply do not reward efforts to improve teaching—teaching is a distraction from what we are hired to do.

Wood suggested a couple of productive directions to take. He was very enthusiastic about a tool physics educators have, called the Force Concept Inventory. It's a simple test that measures student understanding of the general ideas of Newton's laws. It's utility is that it is a way to measure general gains in learning basic concepts in the field. Biology lacks any such tools.

Another useful idea he was pushing strongly was the use of clickers to add more active engagement and collaborative learning to lecture classes (Radagast has been all over this topic.) He explained that they make it easy to add a quick question to a lecture and get instant feedback on how well the class has grasped the material—I can see how after lecturing for ten minutes, it would be incredibly useful to discover that two thirds of the class has no clue what you are talking about before launching deeper into the subject.

What really impressed me, though, is that Wood didn't just tell us about it, he'd handed out the little clickers beforehand, and gave us some examples. After actually using these things, I can see how they would greatly supplement lectures, although I don't think Wood used them quite as effectively as he could have, here. Perhaps because it was a little troubling...

He gave us an example of a question from the physics inventory. It was a simple thing: if an astronaut standing on the moon released a pencil from his hand, would it rise, fall, or float stationary in space? I cringed at learning the scores, and any physicists out there reading this are going to roll their eyes a bit, but a third of this audience of professional geneticists, molecular biologists, embryologists, and developmental biologists got it wrong. Wood waved it off—we aren't physicists, after all—and moved on, but man, I wanted to explore the subject a little more. Shouldn't we expect everyone, even biologists, to understand elementary physics?

The next example was even more disturbing. He offered up a question that he thought was tailored to his audience, that we should all know. The question postulated a simple gene m that had a purely maternal effect, and that, when no wildtype alleles were present, caused embryonic lethality. He had a table with the genotypes of a cross:

The correct answer is that it is alive, of course—in a purely maternal effect gene, all that matters is the genotype of the mother, and that last individual's genotype means that all of its progeny will die, but it itself is fine.

Strangely enough, half the audience got the answer wrong.

Now, I thought that was a result worth pursuing. What does it mean? Are the clickers very poor signal transmission devices with an extremely high error rate? Was there something about the question that was poorly worded, so that even though the audience understood the concept thoroughly, they were misled? Were the attendees so burned out by the deluge of information they'd received in this meeting that their brains were no longer functional? Was the topic so uninteresting that they'd paid little attention to what was asked? Or is the developmental biology audience so diverse that many have no background at all in genetics? That latter possibility isn't at all unreasonable, and I could imagine lots of people could get into development with knowledge of biochemistry or systematics or classical embryology and so forth, where they might not ever encounter the magical words "maternal effect"...but I'd like to know that. That's the kind of information that could be useful when giving presentations to professional groups like this.

But no, Wood just moved on. There was a ripple of murmuring across the room, but the subject was quickly abandoned. That also had me wondering...I was a bit disturbed that something I took for granted may not have been common currency in a large group of my colleagues. Why was I bothered? Why weren't we willing to discuss this? Do I treat the ignorance of my peers with more circumspection than I do that of my students? If the question had been in some basic concept of, say, biochemistry, which was outside my field of expertise, would I also be uncomfortable discussing the matter if I'd gotten it wrong?

The scary thing about these feedback devices, I could see, was that they could easily produce unexpected results, and send your carefully planned lecture skittering down surprising paths. That sounds fun! I was disappointed that this presentation wasn't going to allow that to happen, but the potential is definitely there.

The remainder of the session was good, but not quite as thought-provoking. Phillips described some amazing work he's been doing at Bowdoin, generating online course materials that were tailored to student's preferred learning styles, and using virtual worlds for lab work (students could wander around an island, collecting statistics and life history data from trees they bumped into, for instance—it wasn't anywhere near Halo or Unreal Tournament quality graphics, but the idea was provocative.) Unfortunately, he didn't say a word about how we peons could implement this idea in our classes, so it seemed a bit unapproachable. Klymkowsky ended the session with a description of a project that Wood had mentioned earlier: the development of a biology concept list and inventory. It sounds like it's a year or so away from general release, but they're working on it publicly, at bioliteracy.net.