25 August 2005

Nobody is so blind...

Yesterday, I got my first chance to TA a lab. I've got a nice deal this semester, since I'm assisting a professor instead of running the whole lab section on my own, and because we've only got six students in the lab at the moment. It's not so easy for the students, however. This lab (400-level ecology) is, for most of them, the first time that they are forced to do very much in the way of planning for the experiments. Most of the prior labs that they've taken have been plug-and-chug, follow the cookbook kind of things. Here, they are given a problem, then told to come up with a way to solve it. Yesterday's introduction to the course involved having them measure seed pods from a local invasive (Koa haole), in an attempt to determine which two (of three) bags of seed pods were taken from the same tree. We suggested that sampling would be better than a census approach, and mentioned that two possible parameters to look at might be length of the seed pods and the number of seeds per pod. We also told them to be sure to try to look at everything, as there might be other characters that might work.

Koa haole seed pods tend to hang around the trees for a while, so the bags contained a number of different age classes of seed pod. There were some that were still green, but the bulk were older brown pods. One of the first questions that they came to us with was whether or not the green seed pods were full grown. We looked at them, grinned, and shrugged. They went off muttering to discuss the question among themselves. They were able to quickly reach the conclusion that they didn't know whether or not the green ones were full grown, so they decided to discard them for the purposes of length measurements. They piled up all the green pods from each bag out of the way, then went on to lay out the brown pods to randomly select their samples. The green pods stayed discarded for the rest of the lab, as the students industriously measured lengths and counted seeds.

Since the green pods had been "discarded", the students never looked at them again. Boy, did they miss out. The lab handout didn't mention looking at the ratio of green pods to brown as a possible way to determine which two sets of pods were from the same source, because in past years virtually all of the pods collected were brown. This year, through sheer coincidence, one of the trees that I collected pods from had lots of green ones. The result was that two of the three piles had loads of brown pods and very few green ones. The third had loads of brown pods, along with a whole bunch of green ones. To anyone who wasn't completely locked into a "brown-only" mentality, it was possible to figure out which two piles were similar from the other side of the room. The students, having discarded the green pods, completely missed it. They're going to be spending a lot of time crunching the length and pod number data that they collected, and I have no idea whether or not they are going to find any significant differences.

This is an example of a common phenomenon - so common that it goes by different names. My wife, who is an army flight surgeon, tells me that it the aviators she works with call it "task fixation". I've also seen it referred to as "situational blindness". You wind up focusing so hard on one task that you can completely miss other things that are going on around you. In aviation, this can be life-threatening. In science, it usually isn't that severe, but it still isn't good. It's always important to make sure, when you are counting trees, that you don't miss the forest. Or that you don't miss the guy in the gorilla suit when you are counting basketball passes.

No comments: