Identifying Subtypes Of Beta Bursts In Anxiety-Related Brain Circuits
Andy Bass (2027) | Applied Mathematics
Andy Bass is a SURF L&S scholarship recipient majoring in Applied Mathematics. For his SURF project this summer, Andy is investigating how anxiety-related brain circuits communicate through brief bursts of synchronized neural activity known as beta bursts. Working with neural recordings from the amygdala, hippocampus, and prefrontal cortex, he is using computational and machine learning techniques to determine whether these bursts can be grouped into distinct subtypes. By identifying different patterns of communication within anxiety-related brain networks, Andy hopes to deepen our understanding of how the brain processes anxiety and contribute to future advances in neuroscience and mental health research.
What made you choose this research project? How did your interest in your research topic emerge?
I chose this project because I was looking for a way to sharpen my data science and analysis skills. I had already been working in the Sohal Lab collecting behavioral and neural data from mice, but I did not have much opportunity to explore the data in depth. As I became more familiar with the dataset, my mentor and I began discussing possible research directions. We realized that while techniques such as PCA and clustering are commonly used to identify patterns in complex data, they had not been applied to beta bursts in this context. The possibility of uncovering something new made the project especially exciting to me.
What is your project about (in simple terms)? How does it relate to issues or ideas you have been studying in your coursework?
Researchers in our lab discovered that brief bursts of synchronized activity between the amygdala and hippocampus are associated with anxiety in both mice and humans. So far, these beta bursts have largely been treated as a single type of event. My project asks whether there are actually different kinds of beta bursts that reflect different patterns of communication across the brain. To answer that question, I will examine how the amygdala, hippocampus, and prefrontal cortex interact during each burst and look for meaningful subtypes.
The project draws heavily on skills I developed through my coursework in mathematics, computer science, and data science. Signal processing and machine learning rely on many of the mathematical concepts I have studied, and the project provides an opportunity to apply those tools to a real scientific question.
What are you most excited about this summer? Will your research entail travel, access to special collections, or experiments?
I am most excited by how open-ended the project is. There is real potential to discover something unexpected, and I think the work will open new doors for future neuroscience research. Most of the computational work can be done remotely from Berkeley, but I will travel to UCSF once a week to attend lab meetings and discuss progress with my mentors and other lab members.
What do you hope to learn from this experience? What would success look like for you? What challenges do you anticipate confronting?
I hope to gain experience communicating scientific ideas and working closely with researchers. Whether or not I pursue graduate school, I expect to work in a scientific or technical field where those skills will be valuable.
Success would be pushing the project beyond what I initially set out to do and uncovering an interesting direction for future research. I anticipate challenges related to the size and complexity of the dataset, as well as deciding which features and analyses are most informative. At the same time, even a negative result would be interesting. If the bursts do not separate into distinct subtypes, that would suggest beta bursts may truly represent a single type of neural event, which is itself an important finding.
What do you hope comes out of this research? What impact do you hope your research will have?
I hope the methods I apply prove useful in this context and encourage other researchers to use similar approaches when studying neural data. More broadly, I hope the project contributes to a better understanding of how anxiety-related brain circuits function.
My long-term interest is in neurotechnology that can improve people’s lives. While applications such as restoring speech or vision may still be years away, they depend on a deeper understanding of how the brain organizes and communicates information. I hope this project contributes, even in a small way, to that broader effort.