Scientists as Cheaters

            The topic of “cheating” in science that comes up in the last 70 or so pages of Mendel’s Dwarf brings up an important point to think about when considering scientific discoveries and research. Lambert’s character is looking back and making critical comments about past scientists being completely wrong in their beliefs. They were apparently dim-witted enough to stick to their ridiculous ideas and hold contradictory beliefs to make their theories work. If they had even half the intelligence a scientist should have, they would have realized how mistaken their views were, right? How is it possible that all of the experiments they conducted supported a false idea? Obviously, they must have cheated. But we need to think about this a little more carefully. I believe it was John Carew Eccles who performed experiments for 10 years that proved that transmission between neurons was electrical; but in 1963 he won the Nobel Prize for showing that neuronal transmission is mostly chemical. How can this be? If neuronal transmission is truly chemical, how could a person have done experiments for 10 years in which every single one proved that transmission was electrical? It doesn’t necessarily imply that he “cheated” in some way to make his data work. If you think about it, every lab has a prevailing hypothesis or theory that is the lab’s main point of focus. Right from the start, the lab and its members have a bias and expectation that this is how things work; and all of the experiments they conduct support that view. This isn’t because the experimental results have been altered – it’s because the experiments were designed to support that view in the first place. Who’s going to design experiments to prove their lab’s theory or model wrong? Obviously, you’re going to do experiments that support what you think. And by god, you’re going to stick by your life’s work. For those of you who work in labs, have you ever been in a situation where you got results in an experiment that were different from what was expected? Did your PI, upon seeing this, consider the possibility that the lab’s model may be wrong, or did he/she think you probably screwed something up? Especially in times of shortages in funding, no one in their right mind is going to design an experiment to prove their lab’s theory incorrect. They wouldn’t get funding for proposing something like that. It’s only in times of plentiful funding do people consider conducting radical experiments where they postulate that obtaining a certain result would mean that their model is incorrect and needs restructuring. In times of funding shortages, labs become cautious and stick to what they know.

Mendel cheated. He got results that were so close to his theory every single time, it statistically borders on the impossible. So he fudged his data a bit. He’s definitely not the only one. How often have those of you who do research thrown out an “outlier” in your data? All labs do these kinds of things, and it’s not because they’re purposefully trying to deceive anyone. They truly think that it was probably just experimental error. Sometimes, they’re right; at other times, not quite. Back in the reigning days of craniometry, it was thought that skull size, and thus brain size, reflected on how intelligent a being was; it was discovered – surprisingly – that Whites had bigger cranial cavities than Blacks. And when an outlier happened to pop up? Oh, it’s just an exception – and out it went from the data pool. Even Paul Broca, who is famous for discovering the area of the brain named after him that’s responsible for speech production, was involved in these kinds of studies. Our prior biases and expectations are very important in how we interpret and perceive our findings. We see what we want to see, and we conduct experiments that show what we want them to show. It’s in our nature.

So don’t be so hard on some of those scientists. In retrospect, we laugh at their stupidity. But our time may come, too.

~ B2

P.S. Sorry for all of the brain examples, but it’s what I know.


~ by b2majmudar on April 2, 2008.

2 Responses to “Scientists as Cheaters”

  1. You make a fascinating distinction between fudging on one’s data and designing one’s experiment to support one’s thesis. I suppose if the data ends up disconfirming one’s hypothesis, then the disconfirmation is all the more powerful for arising in an experiment designed to prove the opposite. But what if the results are ambiguous?

  2. Yes, if you get the opposite result of what you expected, then that would be an important piece of evidence. But of course, the problem is that researchers don’t necessarily realize that they’re designing their experiments in a biased way – and they don’t intend to do so at all. So theoretically, getting a different result even when you’ve set things up to get the opposite would be very strong evidence that there’s a flaw in your model. But usually, it’s assumed that something went wrong during the experiment and it takes multiple trials of getting the same result for researchers to seriously consider a major change. If the results are ambiguous, then the experiment will be repeated, possibly with alterations so that ambiguous results are not obtained again. These alterations, though, might make the experiment even more biased toward getting a certain result. So basically, it can be a bad cycle. This is probably why it takes so long for changes to come about and why people who suggest radical ideas are ridiculed for years – until there’s so much data accumulated that supports their theory that they can’t be ignored any longer. I don’t know, is this a very pessimistic view on research? I’ve just been thinking about this lately because part of a project I did for another class involved examining how science/technology can be used to perpetuate things like stereotypes.

    ~ B2

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: