Pseudonymous Ramblings by Psychobunny

Who wants to help me? Please?

I had my first final exam for this semester today, and the first semester of Research Methods (a great class, not terribly difficult, and one I loved but that was completely intimidating) is over! Oh, joy! Oh, rapture! Oh, crapperdoodles!

This means that I only have a month before the second semester of the class starts. In the second semester, you and a partner do your own research, and the IRB proposal has to be finished in the first week of class. I know, then, that if I have the literature review (at least partially) finished and the necessary program written that my partner will go along with my idea. The problem is which idea to use, because I have three I would really like to try and only two chances to do independent research here.

So, please, vote for your favorite in the comments. I'll love you forever, I promise! More details on each below.

One on negative reinforcement absolutely fascinates me. I'm thinking a concurrent fixed-interval 30"--progressive-ratio negative reinforcement schedule (whoa, the jargon!). For the non-psych nerds among us, it means that the participant has 30 seconds to press a button a certain number of times to avoid point loss and that the number of times s/he must press the button increases each time s/he successfully avoids losing points. The twist I want to give this is that the more times the participant successfully avoids point loss, the more points s/he will eventually lose when s/he fails to avoid point loss, so that avoiding point loss actually causes the participant to lose more points than s/he would have if s/he simply failed to press the button at all. The problem? Points are meaningless! I'm having trouble figuring out how to consequate point loss. (Any ideas? I love comments!)

One on resistance to change is really neat, but I can't say much about it. Not many people have worked on it yet, but two Potential Advisors in My Dream Graduate Program recently published a paper on it using animals on a multiple schedule. (My current professors and fellow students don't read blogs. I'm not so sure about these people, and if I end up doing this I will definitely be contacting them to schmooze a bit ask intelligent and relevant questions. I don't want to out myself completely.) I'm an undergrad. I can get away with replications. :) I would like to try it using humans on a mixed schedule. And because it's such a new area, not much literature review to do. I like the sound of that. The problem? When you're giving animals food for responding, you can present noncontingent food as a distractor to test for resistance to change. But again, points are utterly meaningless.

But the one I think would be easiest to get my partner interested in is on design of study materials because it's more neutral with respect to theoretical orientation (i.e., you don't have to be a goose-stepping, Skinner-spouting behaviorist to see the relevance :) ). You know those CDs you sometimes get with textbooks? Some of them have really great study materials on them, but others just suck. What I'm thinking of here is a quizzing program that will repeat the material until the participant achieves 100% accuracy on each of four sets of math questions (each with a different type of feedback) to see which type of feedback minimizes time spent "learning the material". It's also a topic I find interesting since I spent some time this summer writing a similar program for a textbook coauthored by one of Dr. Way-Too-Generous's professors (so I have nearly all the code already written! Major bonus.).

posted by Psychobunny at 12:41 AM