My education has been sorely lacking in the philosophy and history of science. I’m pretty sure that’s the case for most molecular biologists, and probably scientists in other fields as well. Most of us absorb the scientific method over years of exposure, rather than concrete study of the philosophy or history of the scientific method. I’m starting to feel like that’s a major hole in my education that I might gain some benefit from plugging. Everybody’s heard of Kuhn’s ‘paradigm shifts’ (even if we don’t exactly know what they mean), and we all had some “training” in the “scientific method” (visualize air-quotes), but I’m thinking I should take a deeper look at how things really work. Could a better enumerated understanding of the scientific method make generating research questions easier?
I’m increasingly sure that one of, if not the, most important skill for a scientists is learning how to ask questions. Formulating good, easily testable, likely to be successful, interesting questions is something I’ve been struggling with quite a bit. (I’m intentionally avoiding using the word hypotheses because I want to include exploratory, non-hypothesis driven research also.)
Some people seem to have a natural knack for formulating questions, and I’m pretty sure I’m not one of them. I came across this post at Bitesize Bio about the philosophy of science, leading me to some great lecture notes for an intro to biology class explaining something about the scientific method. I wish my professors had been that coherent and had introduced some of those ideas.
The continuum of scientific methods described below (copied from Brandon, RN, Does biology have laws? The experimental evidence. PSA 1996, vol. 2, 444–457.) is an interesting framework for scientific questions. If nothing else it’s kind of fun to try to place the work you’re doing somewhere on the graph.
Description from A blog around the clock:
Collecting the information about all the species of birds and salamanders in the mountains of North Carolina is not a test of hypothesis and is not manipulative (and is not experimental) - yet it is certainly science (place a dot in the bottom right corner of the graph) - it provides important information about the natural world. If patterns emerge from such a survey and prompt new ideas about species distribution, this can then be tested in a more experimental fashion.
Human Genome Project is highly manipulative (and expensive!), yet it is not hypothesis-testing (place a dot in the bottom left corner). Nobody predicted that we would find anything but the four nucleotides known to make up DNA. We had no predictions as what the sequence will be and what would it all mean. Once the work was done, we could use the HGP as a tool for testing new hypotheses, e.g., how many genes do we have, how they are related to the genes of chimps, how diverse are particular gene sequences in human population as a whole, etc.
Paleontology is somewhere in the middle. It is somewhat manipulative (it takes hard work and a lot of people to do it) and it is somewhat hypothesis-testing (place a dot smack in the middle of the graph). Paleontologists do not dig randomly - they dig in particular places on the planet in particular layers of the sediment, looking for fossils of particular kinds of organisms. For instance, a group recently did an excavation in a particular bed of Late Silurian layer, looking specifically for a fossil of an early tetrapod, i.e., a transitional organism between fully aquatic and fully terrestrial mode of life. They discovered exactly that - a fossil named Tiktaalik whose fins were better suited for walking on land than that of fishes (like mudskippers, catfish and lungsfish), yet not completely evolved for land use as in amphibians.
Sometimes nature provides an experiment that tests a hypothesis (a dot in the top right corner). For instance, a biogeographical model of island succession was tested when the volcano Krakatoa erupted and eliminated all life from the island. The scientists went there and observed which organisms flew in from the mainland, in which order, and how the ecosystem passed through several stages until it reached its mature stage, thus confirming (and somewhat modifying) their hypotheses.
I’m still not sure I understand the idea completely. For my own work, at least, I spend a lot of time manipulating data other people in the group have generated and mostly just describing it, while trying to find some patterns that I can generate hypotheses about. I guess it’s the molecular version of collecting information about all the birds and salamanders, but hoping to coming up with a few testable hypotheses along the way.
Does having a better understanding of how the methods you use fit into the larger framework of the scientific process help you to be a better scientist? I’m not all that sure it does for everyone, since several people I know to be good scientists couldn’t care less about the philosophy of science. Yet I have a feeling it could help me. I guess won’t know until I put the effort in to educate myself and find out.
I just couldn’t resist this image from Aminopop reporting on an Economist story about a soon to be published paper finding twins with gay siblings report more partners of the opposite sex. The article suggests that more partners is a valid proxy for reproductive fitness, and therefore that there’s something about the gay gene(s) that make strait carriers more fit.