At the University of California, Santa Cruz, I developed an adaptive Bayesian framework to estimate subjective thresholds and reaction times from psychophysical data.
I also developed probabilistic and likelihood-based models for noisy, partially observed behavioral data, and designed Monte Carlo simulations to evaluate accuracy, robustness, and convergence.
I benchmarked a statistical estimator for median reaction time against a CNN trained in PyTorch. It achieved comparable accuracy while retaining interpretable parameters and requiring no model training.
I built Python and MATLAB analysis pipelines and ran computational studies on UCSC’s Hummingbird HPC cluster. My research has resulted in peer-reviewed publications on reaction-time and subjective-threshold inference.
