During her junior year, independent concentrator Lila Zimbalist was working in Brown’s Curricular Resource Center, advising peers on how to put together an independent concentration. Zimbalist had done it herself, designing a course of study that married her interests in brain science and computer science. At the center, she noticed that a lot of students were coming with the same fascinations about the brain and data and how to combine the two. Seeing the demand, Zimbalist created an unusual senior capstone project – a formal proposal to make computational neuroscience a new Brown undergraduate concentration.
This spring, that proposal becomes reality. By March 16, the first group of sophomore computational neuroscience concentrators will declare, with a number of juniors having already switched their concentration to the new offering. Like the Open Curriculum, the development is classic Brown, another example of students influencing scholarship on campus and paving their own intellectual paths. The new concentration is the first expansion of undergraduate neuroscience at Brown since the original concentration was launched, and brings Brown into a small group of universities, which includes the University of Chicago and the University of Southern California, to offer a focused course of undergraduate study in this interdisciplinary field.
“I am so excited to step into a field that is still defining itself and have the opportunity to grow alongside it,” said Ah-Young Moon, the very first student to declare. “I feel like I’m already beginning to achieve my life’s dream.”
Computational neuroscience is a subdiscipline of neuroscience that brings in knowledge from computer science, cognitive science, applied math, data science and ethics to develop a deeper understanding of the nervous system. Course offerings for the concentration range from neuroengineering to deep learning, from the diseased brain to computer vision, from the history of artificial intelligence to fairness in automated decision making.
Computational neuroscientists can apply their knowledge bidirectionally – using biology to inform computation or use computer science to build models of the brain and nervous system. Because of how few undergrad tracks like this there are, said Carney affiliate Monica Linden, these students will be extremely strong candidates for computational neuroscience graduate programs. If they opt to go to medical school, they’ll become doctors who can immediately grasp the import of computational studies and datasets and what those mean for their patients. And if they go directly into industry?
“They're going to have this computational background. They're going to have used it in an applied way. They’ll have become adept in that interdisciplinary skill of moving between talking to life scientists, talking to computer scientists, talking to mathematicians. I think that makes them incredibly marketable in a lot of jobs,” said Linden, one of the concentration's two faculty sponsors and a distinguished senior lecturer in neuroscience.
Undergraduates have certainly responded. About 30 students were expected to enroll in Introduction to Computational Neuroscience, a new course offered this spring that functions both as a pathway into the concentration or an elective for any student who wants to learn more about the field. Enrollment stands at 68 students.
“Just the fact that we have Carney and the Center for Computational Brain Science right here on campus is huge,” said undergraduate teaching assistant Brian Ji, of the new course’s popularity. “There are a lot of researchers that students have really easy access to.”
The intro course, taught by David Sheinberg, a professor of neuroscience and vice-chair of that department, brings those researchers directly to students. Most Fridays, a Carney-affiliated researcher whose work involves computation guest lectures on their area of focus, which connects to the topics and assignments for that week. Students will hear from the Center for Computational Brain Science’s director Michael Frank on how his lab models and tests neural mechanisms underlying brain processes like working memory, from the center’s associate director Thomas Serre on topics in computer vision, from Steph Jones on how her lab has built software to non-invasively study human brain electrophysiology, and from Carina Curto on how to use mathematical principles to study patterns in neural activity.
“We want students to see the real faces behind this work. The ultimate goal is to get some proportion of these students to say, ‘Can I come to your lab meetings?’” said Sheinberg, the new concentration’s other faculty sponsor, who tirelessly canvassed faculty members during the concentration application to ensure the idea had buy-in and that students would be enthusiastically welcomed into labs.
The design process of both the concentration and the course were deeply collaborative, with Linden and Sheinberg mentoring Zimbalist, Ji and other interested undergrads, advising them while empowering them to take ownership of the projects.
For example, said Ji, Linden taught him and his now-graduated peer Carter Moyer about course learning objectives–the fundamental goals that shape a class–and then invited the two undergrads to write them. Zimbalist, now a graduate working in pharmaceutical consulting with an eye on grad school, described how special it felt to visit the concentration webpage and see the learning goals that she drafted. The students’ contributions are also influencing course and concentration design outside of Brown thanks to presentations delivered at the annual Society for Neuroscience conference.
Sheinberg, Linden, Zimbalist, Ji and Moyer share a vision for the concentration that centers on opportunity. Formalizing the concentration as a Brown approved course of study accessible through student search tools like Focal Point and the University Bulletin introduces everyone to the possibility of exploring this emerging field. The intro course is mandatory satisfactory/no credit, an aspect of Brown’s Open Curriculum that encourages students who may be unsure about a course’s demands to focus on the content and objectives without worrying about their final grade. Sheinberg is very encouraging of those who have no previous computer science experience or advanced math chops, personally working through assignments with students and connecting them with the course’s five undergraduate TAs.
“He preps students for having a toolbox, so if they graduate from this class, they're able to complete a project on their own. That's different from some computer science classes at Brown, which, since they are focused on a specific topic, don't always show you how to start from ground zero,” said Ji.
Ensuring that concentrators think critically and ethically about the ramifications of their research was also important to the concentration’s architects. To graduate, concentrators are required to take at least one course on ethics as it pertains to computer and brain science, and spend a module studying it in both the intro course and in their senior capstone course, which will be offered next year for the first time. These requirements will equip them to play important roles in raising awareness about and identifying biases that could be lurking, for example, in the collection methods of large datasets, or under the hood of AI-powered medical tech.
While the AI boom of the past several years may make this new concentration seem ultra-trendy, Sheinberg sees it as the evolution of a type of neuroscience already prominent at Brown--a transdisciplinary approach he benefited from as Brown graduate student.
“We’re able to bring this concentration to undergraduates now because of a tradition in Brown brain science that goes back to the seventies and eighties of researchers pushing a field without having a degree associated with it, by working at the frontiers of their respective fields,” said Sheinberg. “It’s that spirit that’s drawn such exceptional computational brain scientist faculty to Brown and Carney, and it’s also why faculty who don’t consider themselves to be computational brain scientists will want these concentrators as collaborators. There is almost nobody here who thinks their work wouldn’t benefit from a computational mindset.”