Neuroscience and Music with Poppy Crum
Looking at our larger Stanford community, we want to highlight individuals who are at the top of their fields, and speak with them to find out how they got to where they are. We had the opportunity to sit down with Poppy Crum, Head Scientist at Dolby Laboratories, and do so. Our Editor-in-Chief, Ande Sozzi, was able to speak with Poppy about her passionate work in the neuroscience fields, her personal scientific and musical story, mentorship in education, and to glean any advice she has for young students.
Pixel Journal: When you were growing up, did you know that you wanted to go into the fields you're in?
Poppy: Not entirely. That’s something that I always try to remind people—you don’t have to have it all figured out, but you do have to learn what you’re learning at the time, because it will all come back to factor into your life in ways you can’t predict at any point in time.
Growing up, I was very into music. My undergrad is actually in violin performance. I was also into math, and always took university courses. I think when I was very young—maybe 5 or 6—when I had to write what I wanted to be, I wanted to be a high-tech violinist, whatever that was; so I kind of think that’s what I do.
Where neuroscience factored into things was that it evolved from my desire to solve and know more about the problems I was interested in, and I figured those out as I was in college, as a violinist, studying engineering, and taking elective neuroscience classes. I started to realize that the questions I was asking myself were actually being studied in a field. I would go and spend time in labs on the side, and the mentorship I got while doing that really fostered the understanding that I could study these things, and there were questions to be answered on what I was passionate about.
PJ: You mentioned mentorship—did you have a particular mentor of any kind?
P: I did when I was in music school. I was in Montréal, studying at McGill at the time—this was during my undergrad. I was taking a physics class, and at the time I did a project that was about sound.
A bit of backstory: I have absolute pitch. It’s not necessarily good or bad in terms of being a musician, but your brain does interact with the world in a very different way. It’s like hearing sounds in a similar way to how you see color. I started violin very young, and it was a factor in my having absolute pitch, which I didn’t know until I was about junior-high age—I was in a choir at Interlochen music camp, and they asked me if I have absolute pitch and I said “I know I sing out of tune sometimes, so probably not,” but they told me I did, and I just said “well, cool—I guess I’m different!”
As a musician, having absolute pitch is sometimes not a good thing because it’s much more important, when playing in a quartet, to be listening to whether you want a third flat, or you want it sharp, etc.—there are very subtle differences that you have to be in tune with, who you’re playing with, subtleties of when that piece was written, and how it should come together. But when you have absolute pitch, you hear your own world, and you’re always trying to override that. So one thing I did was play early music—Baroque music is actually tuned to a different tuning system. ‘A’ is something we define, it isn’t an absolute, so A=440—something we’re very familiar with now—was different in the Baroque era. In the Baroque era, A was equal to 415Hz. And for my pitch map, 415 Hz is actually a whole semitone lower, so when I play Baroque music and I play what I think should be an A on the paper it sounds like a G sharp. And I was like “Oh my gosh, I’m terrible. Nobody will want to play with me constantly having to recalibrate my mind” But at the same time, I was in an elective neuroscience class by a McGill professor, and we read a paper by Eric Knudsen, who is actually a professor at Stanford. The paper was about auditory and visual neural maps of space; how you form aligned neural representations for a particular location for what you see and what you hear. So to understand the dominance between vision and audition, what he did was rear owls with prism glasses on that would shift their visual world by, say, 15 degrees, and then put them in environmentally relevant and pressure situations where they had to hunt with these prism glasses on. It forced the system to be plastic and adaptable, and what he saw was the neural connections of these maps realigning. You would develop new and different maps. At the same time I had been playing my Baroque music at A=415, and 6 months in, I realized I had absolute pitch at 415 as well as 440 Hz—I had developed a secondary pitch map, and now I could play in tune again! So I have these two pitch maps and nothing in between, and that made me realize that my brain was a little different, and inspired me to really want to understand more problems about neuroscience and perception.
At the time I was taking a physics class, and was doing my final project on acoustics and perception, and I told my Professor that I really wanted to know more about this sort of research, and he advised me to talk to Dr. Albert Bregman in the psychology/cognitive science department. I went and knocked on his door, not really knowing who he was—his past research or well established and respected reputation. I was just interested in the problems, and excited to talk to someone who I was told could help me understand how to think about these types of questions more scientifically.
Albert Bregman defined a whole field, and his work is seminal to many problems we solve here [at Dolby], and to many of us. And he was so receptive to someone coming and wanting to ask thoughtful questions. He said “why don’t you come work in our lab on the side,” and so I began doing so, and he was a great mentor. I would go sit in his office, and just discuss philosophical problems, and he really taught me how to think, and how to write in a scientific perspective that is invaluable. There are a lot of phenomenal writers who have worked with him as well, like Steven Pinker, and Allison and Adam Gopnik among others.
There was recently an article that Adam Gopnik wrote in the New Yorker, and he attributed a quote to Al Bregman. I guess [Adam] did a similar thing, and would go sit in Al’s office and chat about life. He quoted Al when he was talking about his life decisions as an undergrad: ‘Do I go into arts and writing, or this direction, or this direction—I don’t know, and it’s very complicated for me.” Al said, “Well, if it’s a hard decision, it doesn’t matter” because there are good reasons for both. I use that line now.
PJ: How did you come to Dolby?
P: Ultimately I became a neurophysiologist, and spent many years studying the brain and studying single cells, treating the brain as a cortical circuit—trying to understand how we can go from neuron to neuron in what we see and record in the brain, to experience and perception. My specialty is sound, but also how what we see and hear interact (sensory integration)—thinking about how the brain does that, how we model that in an algorithm, and how we think about how that leads to experience.
Well, Dolby is a place that thinks about sensory perception, and has been doing so for fifty years. Years ago, I knew about Dolby when I was a musician because I always worked as an audio engineer, working for radio stations and things like that on the side because I was passionate about it and interested in it. I made records to make extra money—things like that. The way [Dolby’s] technology worked didn’t just solve the problem from an engineering perspective, but it had an understanding of human experience and what we call psychophysics, or algorithms that took into account human perception. So, say you think about an MP3, or any perceptual compression algorithm—such as Dolby Digital – that’s what you hear when you watch HD TV. These algorithms take a lot of information and reduce it down to a small size, but do so not just mathematically, but also with regard to an understanding of human perception. You allocate information towards what someone experiences and away from what they don’t– it’s all about how you weight the information. You can make a sound or video file a lot smaller computationally without changing what someone actually hears or sees.
Dolby came and recruited me when I was Research Faculty in the Department of Biomedical Engineering at Johns Hopkins University. I was a primate physiologist when I was there—I worked with monkeys for many years. The reason I was interested to come [to Dolby] was that the opportunity here was so large; change was happening in computation and new technologies, and where that future was going—there was this big, open, wide opportunity in the area of thinking about sensory experience and processing, which didn’t used to be as rich because of where computation was. There was this opportunity to really think more about how the brain solves problems, how we experience the world, and how we develop technologies that are intelligent and modulating towards our worlds, and are in tune with current sensory states. Dolby, as a company, really thinks about what we hear and see—many of our most innovative technologies right now are actually in image processing—and so it was very enticing to think about these possibilities.
Throughout my whole time as a scientist I also stayed connected to my past passions—music and sensory technology—and this is something I would recommend for all people—you’re often in a field working with experts, collaborating with experts in your area, and to be in that world, you often have to be very narrow. But I also tried to continue doing lectures outside my immediate field-–to stay grounded to the broader relevancy--and speaking about the understanding and interpretation of neuroscience and perception to the entertainment and music communities, and so that’s how Dolby was aware of me. I also kept performing as a musician--critical for me to stay grounded to things I enjoy and how all the points along my trajectory fit together.
PJ: You were a fellow of the US Defense Science Research Council—what can you tell us about that?
P: That was an incredibly interesting group of people and scientific insight. That group, and there are other efforts like this, was a think-tank of experts that span physical sciences, social sciences, biological sciences, and technology. Sensory neuroscience--my specialty--I think became more relevant because of things happening in robotics and, again, computational capabilities. I was invited to share a presentation to the group, and then invited to become a fellow. This particular group had multiple academic and industry members. Those were some of the most brilliant minds I have worked with. The efforts of the members of that group have had a huge impact in many scientific areas over the past 40 years, and it was both humbling and inspiring to interact with the breadth and depth of the collective.
PJ: You are a US Representative to the International Telecommunications Union--what is that like?
P: That’s a part of my job here [at Dolby]. The International Telecommunications Union (ITU) is a specialized agency of the United Nations (UN) that manages international issues regarding information and communication technologies.[ I work with multiple standardization groups that support the efforts of the ITU and meet in Geneva a few times a year. The reasons these types of organizations exist is so that there is an international hub where delegate members from all countries come together to establish interoperability standards so that the world can be connected, and so that the world can share information successfully and in ways that are not going to bias one country over another, which has a lot of impact on many different countries in the way that information is represented.
I also do other things in that realm. For example, I co-chair an organization for the Consumer Technology Association, which is quite interesting. That group specifically considers wearable technologies and how these technologies can provide amplification to users. It’s interesting because it’s at the grey-area intersection of where current technological device capabilities wander between consumer and medical classifications. I’m somewhat active in trying to change the US regulatory standards on hearing aids and hearing enhancement. It’s a place where the US is very different from, say, Europe. Right now [the US] has more than 30 million people that suffer from hearing loss, and much fewer than 10 million (and fewer yet in financially restricted and demographically challenged areas) ever get hearing aids or anything that can mitigate their hearing loss. What we have is a situation where hearing aids aren’t covered by Medicare, and they’re exceptionally expensive even though the technology in them has gone down in cost. Right now it is the situation that companies can build technology that can help someone with hearing loss for much less, but they can’t actually market these devices for what they are unless they become medical devices which can have a huge cost overhead and entirely different distribution pipeline. I’ve testified to the FDA, and spoken to Obama’s taskforce on science and technology. The Institute of Medicine has issued a call for change which is a substantial milestone to all hoping to see this happen. At this point in time we are hoping for change in that area so that more people in the US can benefit from the technological innovation and current capabilities that exist.
PJ: What courses do you teach at Stanford?
P: I teach a course I created which is called Neuroplasticity and Musical Gaming, but it’s more about the power of technology and controlled interactive environments to bootstrap learning and neural plasticity in ways that we can try to predict. It also tries to leverage the fact that I’m in a position to think about the latest controllers and embodied environments - to use those tools—whether it’s VR, or gestural control, or whatnot—to create environments that are the most natural for whatever the learning point is. There is also a fairly heavy biological component and coding component to the class. It heavily thinks about sensory interaction and the impact that one sense can have on another, and amplifying learning. The course involves students coding multiple targeted games throughout the quarter with a final project aimed towards learning and neural plasticity in a topic that resonates with their individual passion. For example, I’ve had highly competitive athletes and musicians in the course who build games really trying to focus on what helps define their unique super powers in a neural game environment. I really love how different every year is! I also supervise students for honors theses and things like that.
PJ: Do you have any advice for our readers who might be interested in a particular scientific field?
P: Two things:
1) I think mentors are very important. I don’t think you can start early enough learning and finding people to talk to—finding the experts and approaching them, and going to conferences. Look for who is doing interesting things. Spend time. You won’t necessarily know who you want to be your mentor. My perspective is that if you’re going to go to grad school, or you’re going to go to an undergrad program where you want to do research, go there because you already have interest in who the people are and how they think, and know there are ideas you find interesting, because it is just must more fulfilling to be chasing ideas than to be in a program where things are prescribed to you. Even though those [ideas] will completely change, that interaction at that point in time is very positive.
2) Don’t discount anything you’re learning, even if you think it’s not relevant to where you want to be. I can remember, years ago, learning how to solder circuit boards. I worked for $10 an hour just to solder circuit boards for one of the inventors of the synthesizer so that I could be exposed to his ideas. In addition to all the memorable conversations and experiences working for him, all that soldering came back for when I needed to fix wires and build parts of my lab—later it wasn’t a block, but, instead, it was part of my toolset. You don’t know what the future is going to be, and don’t limit yourself in trying to predict that too early – enjoy the ride.
PJ: Thank you so much for your time, Poppy!
Find Poppy on LinkedIn, here!
