Chasing the Mystery: A Data Driven Look Into the Biological Basis of Psychosis at the UNC Neurocognition and Imaging Research Lab - with Samhitha Pudipeddi

Hello! My name is Samhitha Pudipeddi, and I am a junior at UNC Chapel Hill majoring in Neuroscience and Computer Science with a minor in Chemistry.

Growing up in Indian culture, mental health and psychiatric disorders were not topics that were openly discussed. Since conditions like depression or ADHD didn’t have visible symptoms, like a rash or a broken bone, they were often seen as an illusion or an excuse. However, once I started working as an EMT, I saw firsthand how devastating psychiatric conditions are on individuals and their loved ones. Call after call, I witnessed families torn apart by addiction, patients struggling with psychosis, and individuals battling years of depression and hopelessness. Although we had established protocols to treat conditions like hypoglycemia or asthma, we lacked clear treatment plans for psychiatric emergencies. Beyond sedation or transport to a facility, we had few tools to address the root of the problem. While we have made massive strides in understanding the heart or the lungs, the brain, and the disorders that affect it, remain deeply mysterious. As they say, we know more about the depths of the ocean than we do about the human mind.

That mystery drew me in.

As a freshman, I joined the Christoffel Lab, where I worked on mapping the neural circuits involved in hedonic feeding using a mouse model. It was my first look at how the brain drives behavior, and I loved it. Around the same time, ChatGPT and large language models exploded into the world. I was in awe of how software could mimic thought and act as a primitive model of the human brain. As I started to explore this topic, I learned that uncovering patterns in large datasets, something I had already been doing in my neuroscience research, was also the foundation of how artificial intelligence systems like ChatGPT function. This connection fueled my curiosity about how data analysis could be used to unlock the brain’s mysteries. As I studied computer science, I became excited by the idea that code and computation could help us get closer to the root of psychiatric illnesses.

When I shared these interests with the Gil Internship Program, they placed me in the perfect worksite.

This semester, I interned at the Neuroimaging Research in Psychiatry Laboratory (NIRL), founded and directed by Dr. Aysenil Belger. The NIRL is dedicated to understanding how attention, emotion, and executive function are represented in the brain, and how these processes are disrupted in psychiatric conditions like schizophrenia and PTSD. Using tools like fMRI, EEG, and physiological stress markers, the NIRL seeks to discover the brain’s hidden signatures, with the ultimate goal of designing more effective, personalized interventions.

​​What made this experience so exciting for me was that I got to work with real human data for the first time. My project focused on adolescents with a family history of schizophrenia, a population that may already show subtle stress regulation differences before any clinical symptoms appear. I specifically looked at cortisol responses to acute stress and structural differences in the hippocampus and amygdala, two brain regions involved in regulating the stress response. Since the data was collected already, my work was focused primarily on analyzing the data.

Each t-test and ANOVA I conducted wasn’t just a statistic: it was a clue. Using R, I parsed through hormone levels, brain volume measurements, and heart rate data, searching for patterns that could help explain why some individuals are more vulnerable to psychosis than others. I was also given the opportunity to work with multiple departments during this internship. Dr. Belger guided me through the neuroscience behind our hypotheses, while Dr. Kinh Truong from the biostatistics department helped me test them. Just from the few variables we tested from the study, we found that adolescents at high risk for psychosis released significantly less cortisol in response to an acute stressor compared to adolescents at low risk of developing psychosis. Excited to find more in the data, I will be continuing this project next year for my senior honors thesis.

I am incredibly grateful to Dr. Belger for her mentorship and to Dr. Steven Buzinski and Emily Dolegowski for making the Gil Internship Program possible and for creating a space where students like me can explore their passions. As an aspiring physician, this experience has not only sharpened my research skills, but it has given me a deeper understanding of the people I hope to serve. I will carry everything I learned at the NIRL with me as I continue my journey toward becoming a physician and a scientist.