Luck and Potential

Recently, Steven J. Luck answered a few of our questions about his work on event-related potentials. An ERP (to use the insiders' lingo) is a measurable electrophysiological response to a thought or mental perception. Steve's book on ERPs was published in August, and below he discusses how he came to the topic while working in the lab of Steve Hillyard, of UC San Diego, and how the experience in the lab directed his own research.

MITP: What would you like potential readers to know about your book?

SJL: It's really intended to be a how-to book. It's not quite "ERPs for dummies," but my goal was for it to be a practical guide to designing experiments, recording high-quality data, and analyzing the data. But it doesn't just provide a cookbook -- it also explains the reasons why things are done in a particular way.

MITP: As I understand it, you began working in earnest with the event-related potential technique in the late 1980s and early 90s in Steve Hillyard's lab at the University of California, San Diego, which had a long history as an electrophysiology-oriented lab. What was working in that environment like?

SJL: This was an amazing environment for a young graduate student. I didn't fully realize it at the time, but I was constantly surrounded by people who had shaped and who have continued to shape the field of ERP research (and, more broadly, the emerging field of cognitive neuroscience). The combined wisdom of these individuals was most commonly shared at our weekly lab meetings, which started promptly at 4:07 every Monday afternoon and lasted a full two hours. Each week, someone would give a detailed presentation on his or her ongoing research, and every detail of the experimental design and the data were examined carefully by the group. These lab meetings were very intense, and it was a little bit intimidating to be the speaker. But no one was ever nasty, and it was a very supportive environment. Much of my book is based on what I learned in those lab meetings.
    Another key aspect of the Hillyard lab during that period was the combination of technical expertise and experimental design emphasis. Steve Hillyard is an extraordinarily careful scientist, and he fostered an environment in which everyone strove for experimental designs that were both clever and airtight. This was also emphasized by the other grad students and postdocs in the lab. Marty Woldorff, in particular, had an almost uncanny ability to find subtle confounds and alternative explanations. Steve Hillyard was not, however, very technically inclined, especially when it came to computers. But Jon Hansen was truly brilliant when it came to technical matters, and he was the lab's technical guru. He designed the lab's hardware and software, and I learned a great deal about signal processing just from reading his software manuals.
    The Hillyard lab was also somewhat unusual for an ERP lab in that it was not completely focused on ERPs per se. Steve Hillyard was an attention researcher just as much as he was an ERP researcher, and our attention research was designed to answer questions that were central to the field of attention research. Similarly, Marta Kutas and Cyma Van Petten began doing language ERP studies that were very sophisticated from a psycholinguistic perspective. The goal of the research in the Hillyard lab was to have a broad impact on cognitive scientists and neuroscientists, even those who didn't care anything about ERPs.

MITP: How has the experience of working in the Hillyard lab affected the direction of your research?

SJL: Although no one ever made it explicit, an important characteristic of the Hillyard lab was that it was easy to run experiments, and we therefore ran a lot of them. Hillyard and Hansen made sure that we had access to powerful and flexible computer systems that could be rapidly configured to conduct and analyze almost any experiment we wanted to run. Grad students didn't need to be computer programmers, and we could usually get an experiment running in a few days, collect the data from a dozen subjects in 2-3 weeks, and then complete the data analyses in a few days.
    When I started my own lab at the University of Iowa in 1994, I was determined to implement everything I had learned in the Hillyard lab. Most importantly, I wanted my students to conduct perfectly designed and flawlessly executed experiments that made a broad impact on the field of attention research. Perfection is hard to achieve, but we inherited the Hillyard lab's software systems, so we could easily run and re-run experiments until they met our high standards. I became convinced that it was best to use simple, time-tested approaches so that we could conduct experiments rapidly rather than spending countless hours trying new-fangled approaches that had not been validated.

MITP: Is there any aspect of running a lab you find especially rewarding?

SJL: My favorite part of the research process is designing experiments. It always has been. And I especially enjoy doing this collaboratively with my students and colleagues.