STEPHEN ENGEL & RUSSELL POLDRACK
Ask good questions
Exercises & assignments
Work with real data
Designing Experiments with Functional MRI
Just the notion that the data they were working on was from their brains was exciting for them. Instead of working through Excel sheets, they were looking at pictures of their brains while they were doing something interesting.
BW: How would you describe your project?
SE: The basic idea was to teach a lab course in psychology where people learn the basics of experimental methodology and how to do good science. We try to make it exciting by teaching them to work with functional MRI; they design psychology experiments that measure brain activity while subjects perform various tasks.
RP: They learn the skills and the techniques for designing experiments and interpreting data from those experiments, skills you need in order to do any kind of science. The difference between the standard course in cognitive psychology and the one that we teach is that this one uses the cutting-edge technology that's being used in the field.
BW: Where and how did your project begin?
SE: When I came to UCLA, another expert in this field and myself had the idea of teaching an undergraduate lab course incorporating functional MRI. It's an advanced topic that people usually get into in graduate school, or even later. Eventually we got support from the psychology department, and we're very grateful for that. When it came time for actually doing it, Russ had shown up on campus, and he seemed like a natural participant. Initially, we were motivated by the challenge of trying to bring people into the field. We wanted to get people interested at an early level. Since then, the field's grown exponentially, so that isn't such a motivating factor any more. But in the future we'd like to work towards getting high school kids involved in our discipline.
BW: Can you describe how the technology is incorporated into the course?
RP: We start by introducing them to the concepts behind the technique. The technique uses magnetic resonance imaging, which is a relatively new method of imaging the insides of people's bodies, non-invasively and without injecting anything. We spend the first part of the course teaching the students about the kinds of data the technique yields, and how to interpret those data. Then we jump right into how to design experiments that use the technique. Just as you might use the Skinner Box or some other piece of equipment, we use the MRI as a jumping-off point for designing experiments, thinking about how to ask a question, then once you've done the experiment, how you interpret the data to determine what the answer to the question is. It's a fairly complex procedure--we spend several weeks talking about the different aspects of data analysis, which involves both processing the signals that come out of the scanner and running statistics to figure out what parts of the brain are active when the subject is performing particular tasks. They spend a good deal of time doing exercises with canned data that teach them how to do the various operations involved in analyzing it.
In the latter third of the course, the students break into groups and design experiments that they run on the MRI scanner. The last time we did it, there were four groups. One of them was looking at different visual processes; another was looking at the differences in activity when trying to remember words in Spanish versus English. Whatever question they're interested in, they come up with an experiment to ask that question.
There's usually one person in each group that becomes the leader and pushes the agenda, and that's fine. At first they sit together for a couple of sessions talking about what their experiment will be, and then with our guidance they come up with the specifics of the design. When it comes to running the experiment, it's generally computerized, so that when the person is in the MRI scanner we can use the computer to present stimuli. For example, we might show words or play sounds to the subject in the scanner. We help the students implement the computer programs to run the experiment. When we take them over to do the scanning at the Ahmanson Lovelace Brain Mapping Center, they are each involved in collecting the data.
Once the experiments are complete, we load the data onto computers in one of the labs over in psychology, and they walk through each of those operations. They divide up the tasks involved according to their strengths, so each member of the team contributes to the data analysis. They each have to write a paper, but they present their experiment to the class as a group.
SE: The course is still evolving. We've taught it twice now, and I think it's gotten better each time. We've scaled it up to teach more students, and we're hopeful that this award will help us to go even further with it, so we're grateful for that. It's continued to evolve in complexity, and we're constantly trying to be more efficient in covering the technological side of it.
BW: Were there any surprises? Did they take it in directions you didn't expect?
RP: Certainly, in terms of the kinds of questions they wanted to ask, they went in directions that I don't think we would have ever imagined. They were very creative, and worked off of what they had learned in their previous psychology classes. There were a few experiment ideas that were just too far out there, or required methods we couldn't implement. But in general, we were really impressed with the creativity that they brought to this task. I think they all came out of it feeling like that had really produced something. For many of them, it was probably the first real experiment that they'd ever run in their lives, where they came up with the idea and actually collected the data. When it works, it's satisfying, and for all of the groups that did experiments this year, it pretty much worked.
BW: What has been the students' reaction to the use of this technology?
RP: There's been a range of reactions. I think most of them were excited by it. They have to learn a lot of difficult material to be able to understand these things, and some are in a better position to do that than others. But even if they don't fully understand the MRI part of it, they understand how to design experiments and how to interpret the data. Many of them get really excited and end up working in one of our labs doing research once the course is over.
SE: Just the notion that the data they were working on was from their brains was exciting for them. Instead of working through Excel sheets, they were looking at pictures of their brains while they were doing something interesting. I think that really helps to motivate them. For a lot of people it's eye-opening, also, in that they don't expect psychology as a discipline to be this high-tech. They're in psychology because they want to be a therapist or a clinician, and their technology's a note-pad and pen. To see this side of it, where we're using this fancy medical equipment to take these images of peoples' brains, really showed them what the field is and could be. Working with the images gives you the feeling that the data's real and so you can more easily visualize what's happening and understand it. Personally, I like that about the field, and I think a lot of the students get that feeling as well.
BW: Does this project energize them for the rest of the course?
RP: They actually do their experiments towards the end of the quarter, and I think a lot of the excitement arises from their awareness that they get to go do that. Certainly once they're finished and have written their reports and presented them to the class, a lot of them want to do research with one of us, or to take further courses, to learn more about these techniques.
SE: The practical experiment methodologies we teach in the course are clearly very important--in some sense more important because you really have the potential to waste a lot of money, effort and time using these fancy technologies, if you're doing stupid things with them or if you're designing bad experiments. So what we try to do in the first half of the course is not only point out what you can do, but show them a lot of examples of things you really don't want to do. It's definitely our hope that using this exciting method will motivate them to learn this material. When they put their friends in the MRI scanner, the stakes are actually quite high, and they can imagine the immediate consequences of a badly-designed experiment because they're so involved in the work. So I think they are motivated to get through the basic material with a good understanding.
BW: Was it worth the effort?
SE: Yes, I think it definitely was worth the effort. Psychology is a very large major, and so in a lot of our classes we're dealing with hundreds of undergraduates. To have a smaller group, and to be able to work with people closely and see their enthusiasm really was a big payoff, both in terms of motivating me for teaching, but also, for research, because when you see them so excited by this technique and what it can do, it feeds back into your own enthusiasm. Teaching what you love to small groups of motivated people--it just doesn't get any better than that.
RP: I agree that it's been rewarding with respect to both research and teaching. We've seen students who have come through this course come to our lab very excited about doing research. Because this 121 course is something they take fairly early in their careers, if we can excite them at that point, they have a lot of time to really get into the research before they graduate. It's nice to be in one of the few places in the world that actually teaches this technique to undergrads. We have one a course here for graduate students as well, but teaching it to undergraduates is something that I think a lot of people have assumed is just too hard, because there's too much to learn. We've shown that we can do it with the help of these technological tools.
SE: We feel like this course offers a unique undergraduate opportunity that you can only get at a university that's really top-notch in both research and teaching, like UCLA. By bringing in this cutting-edge technology into the undergraduate curriculum, we give students an experience that they're not going to get at a small liberal arts college. That's really valuable for undergraduate education.
BW: What's in the future for this course?
SE: We're going to continue improving it. Part of that involves improving our guidance of the students in designing interesting, practical experiments. We also hope to make the introductory materials more topical and interesting, to smooth out the lab exercises before they get to their big final project. Beyond that, we're hoping to present the course at the International Neuroscience Conference in the fall. They have a section on educational techniques for neuroscience and we're pretty hopeful that people will be interested in this, so that maybe the course will spread to other universities.
Oral Interview by Bill Wolfe, March 2004