Research
Exploring the intersection of computational neuroscience, biomechanics, and artificial intelligence through rigorous academic research
May 2024 – Jul 2024
Georgia Institute of Technology
Undergraduate Research Assistant
Biomechanical Simulation & Motor Control
Advanced computational modeling of feline motor function and biological neural networks, focusing on the intersection of biomechanics and artificial intelligence.
Key Contributions
- Maintained and upgraded MATLAB-based biomechanical simulation models for feline motor function
- Implemented optimizations that improved computational efficiency by 15%
- Developed ETL data preprocessing pipelines using Python and MATLAB for neural network models with 10,000+ samples
- Analyzed computational architectures comparing artificial neural networks to biological motor control systems
Feb 2024 – May 2024
Georgia Institute of Technology
Research Project (Coursework)
Neural Language Encoding Models
Implementation and analysis of semantic encoding models based on Huth et al. (2016) to predict brain responses to natural language stimuli. Built regression models using word embeddings and Finite Impulse Response (FIR) to map semantic features to BOLD fMRI responses across cortical voxels.
Key Contributions
- Implemented encoding models using English1000 word embeddings (985 dimensions, 10,470 features) to represent semantic content of naturalistic story stimuli
- Built FIR models to predict hemodynamic responses across 2-8 second windows, achieving significant prediction accuracy for semantic-selective voxels
- Analyzed 20+ best-predicted voxels to identify semantic selectivity patterns, revealing distinct cortical representations for court-related, color, and attribute categories
- Compared semantic vs. phonemic encoding models via correlation analysis, demonstrating superior performance of semantic features in predicting language-responsive brain regions
- Processed and downsampled stimulus representations to match fMRI temporal resolution, handling high-dimensional feature spaces across multiple story datasets
Jan 2024 – Mar 2024
Georgia Institute of Technology
Research Project (Coursework)
Brain Region Classification Using AI Models
Computational vision neuroscience research utilizing Murty Lab's virtual Ventral Temporal Cortex (vVTC) AI model to identify and characterize brain region selectivity. Systematically tested three visual processing regions using diverse stimulus categories to understand neural response patterns and model limitations.
Key Contributions
- Successfully identified and classified three brain regions (FFA, EBA, PPA) using AI-based predictive model achieving high accuracy across 55+ test stimuli
- Investigated pareidolia effects by testing face-like objects, revealing model limitations in processing optical illusions compared to biological neural responses
- Discovered region-specific selectivity patterns: FFA preference for conventionally attractive facial features, EBA heightened response to human digits, PPA selectivity for tree-containing landscapes
- Identified edge cases that 'fool' each region, providing insights into computational differences between artificial models and biological visual processing systems