Caveat lector: This blog is where I try out new ideas. I will often be wrong, but that's the point.

Home | Personal | Entertainment | Professional | Publications | Blog

Search Archive



SciPle.org interview

Lots of interviews this week!

Recently I presented some of the latest developments in my brainSCANr project at the Society for Neuroscience conference.

At my poster I was approached by SciPle.org, and Leonardo Restivo asked if I would do an interview with them. Well the interview's been posted and my answers are below. They're more personal than I was expecting, but I try not to shy away from personal questions (even if the answers can be uncomfortable).

Check Sci.Ple out; follow them on Twitter. They've got some really cool ideas that I'd love to see work out.

Sci.Ple: What is your background?
Technically I began my undergraduate career at the University of Southern California as a physics major. I grew up in San Diego with a lot of clear night skies. I wanted to be an astrophysicist. But that was not to be my path, and I ended up doing a degree in psychology while taking a host of philosophy and cognitive and computer science courses. I then worked as a research associate at UCLA for Edythe London; I was the PET scanner operator and radioactive pee cleaner (that's another story). In 2010 I completed my PhD in neuroscience at UC Berkeley under the mentorship of Robert Knight, the institute's director at the time. I'm currently a post-doctoral fellow at UCSF working with Adam Gazzaley.

Sci.Ple: Among your published papers, which one is your favorite?
Okay, I'm cheating a bit since it's not actually published yet, but it's close so I'm just going to act like it's published and hope that talking about it doesn't jinx the whole process. My favorite paper is one I co-wrote with my wife Jessica and is titled "Automated Cognome Construction and Semi-automated Hypothesis Generation". If I play by the rules of the question, then I'll go with "Hemicraniectomy: A new model for human electrophysiology with high spatio-temporal resolution" (Voytek et al., J Cogn Neurosci 2010).

Sci.Ple: Why is it your favorite?
Honestly I believe that the "Semi-automated Hypothesis Generation" paper is a Big Deal. We're text-mining the abstracts of millions of peer-reviewed neuroscience papers to try and make some sense out of how neuroscientific concepts interrelate. Then we're going one step further to see if we can find statistical "holes" in the literature… we're literally trying to (semi)automate one aspect of the scientific method: hypothesis generation.

Sci.Ple: What was the most challenging part of this paper?
The first spark of the idea for this paper came about during a panel I was on at a cognitive science conference in 2010. In response to a question from an audience member, I said something like, "the scientific literature is smarter than we are; many basic facts about brain function are probably already known, but we suck at synthesizing it all." Trying to prove that statement was the foundation for his paper. In order to write the paper I had to learn about text-mining, some graph theory, etc. It's totally outside of my comfort zone, but I think it's too cool to stress out about it not being "perfect".

Sci.Ple: What drives you in your day-to-day job?
Honestly? People literally pay to me to think about cool shit. That's my whole job. I get to say "I wonder if anyone has done this before," and then go out, run some experiments, and then possibly learn something that no one has ever known before. I've worked on a loading dock carrying heavy things onto trucks for up to 16 hours at a time. I've worked at a motel where my job was to go room to room and collect all of the... soiled... linens. So I guess the simplest answer to what drives me is a mix of awe and perspective. That's what keeps me going to work every day.
Sci.Ple: What is the most exciting part of your job?
Talking to other people. Collaborating and combining the accumulated knowledge of multiple brains in new ways to tackle hard problems. That's amazing. It's humbling and inspiring to see brilliant people's minds work.

Sci.Ple: The least exciting?
Paper formatting and data munging.
Sci.Ple: Name a scientist whose research inspires you
Reed Richards. That guy seems to be able to produce an endless stream of amazing, breakthrough ideas. Oh. Seriously? I've got a huge amount of respect for pre-digital scientists. Lots of them were going out and just trying everything. Sever the nerves in your own arm? Why not?! Cover yourself in varnish to prove that the skin does something important? Who cares if you almost died? You were right! I don't think I have the… fortitude… to do a lot of what those folks did--from the crazy self-experimentation to the drudgery of pre-digital writing and researching--I can't help but respect that.

Sci.Ple: What are the next frontiers in neuroscience?
Information integration. How do you go from a neuron giving off an action potential to a thought? From a neurotransmitter binding to a receptor to art? Imagine a Venn diagram that contains medicine, biology, genetics, psychology, philosophy, engineering, computer science, mathematics. That's neuroscience. Knowledge in each of these fields is being slowly accumulated (almost always separately) by doctors, biologists, geneticists, psychologists, philosophers, engineers, computer scientists, and mathematicians. We need more cross-disciplinary work.
Sci.Ple: Why science?
Why anything? My life, like almost everyone else's, has consisted of a series of half-blind stumbling steps. I got lucky. That said, I'll take the minor frustration of getting a paper rejected or having someone disagree with me over "soiled" towels and sheets any day.

Sci.Ple: If not science?
I'd work as a bartender at a pub. Or try and run my own. I love meeting and talking with people, and few places are better for than that a nice local watering hole.

Sci.Ple: Why?
"What happens to a werewolf on the moon?" I read that online the other day and it still cracks me up.