pearl franz
ai, neuroscience, and biomedical research projects
Embedded Files
Neuroscience Research: Linguistic Connectometry
A study exploring the relationship between white matter tracts and linguistic performance in healthy adults.
Neuroscience Research: Linguistic Connectometry
A study exploring the relationship between white matter tracts and linguistic performance in healthy adults.
Fiber Tractography Lab, University of Pittsburgh
In this project, I used diffusion MRI and connectometry analysis to investigate how individual differences in language processing relate to structural connectivity in the brain. Specifically, I examined correlations between white matter integrity and performance on linguistic tasks such as picture vocabulary recognition and reading comprehension.
Using tools like DSI Studio and quantitative anisotropy-based tractography, I analyzed over 1,000 dMRI scans and identified statistically significant tracts associated with syntactic processing across individuals. Findings revealed distinct connectivity patterns associated with linguistic performance, including increased connectivity in the arcuate fasciculus, inferior longitudinal fasciculus, and uncinate fasciculus, and decreased performance associated with greater connectivity in regions of the default mode network, such as the corpus callosum and cingulum. These results suggest a structural basis for individual variability in language comprehension.
This work contributes to our understanding of how microstructural differences in brain connectivity support complex cognitive functions like language comprehension.
Biomedical Engineering Research: Transabdominal Fetal Pulse Oximetry
A study on isolating fetal pulse signals from the maternal abdomen using signal processing methods in MATLAB.
Biomedical Engineering Research: Transabdominal Fetal Pulse Oximetry
A study on isolating fetal pulse signals from the maternal abdomen using signal processing methods in MATLAB.
Biophotonics Lab, Carnegie Mellon University
For my research on signal processing for transabdominal fetal pulse oximetry, I researched how best to isolate the fetal signal from a mixed maternal and fetal signal. This involved the filtering of both real animal data as well as a Monte Carlo stimulation, and comparing the efficacy of traditional signal processing methods (e.g. Fourier Transform, bandpass filtering), as well as investigating the efficacy of Principal Component Analysis and Independent Component Analysis.
Artificial Intelligence Research: Computational Model of Human Learning
An investigation into memory retention patterns in an AI learner using human-like decay simulations.
Artificial Intelligence Research: Computational Model of Human Learning
An investigation into memory retention patterns in an AI learner using human-like decay simulations.
Human Computer Interaction Institute, Carnegie Mellon University
As a researcher for Apprentice Learner (AL), a computational model of human learning, I studied AL's ACT-R memory implementation and compared the memory retention of AL to that of humans. More specifically, I coded JSON files that would stimulate spaced repetition and cramming conditions. I found that AL was able to demonstrate the recency effect, but not the primacy effect that is typically found in human memory retention studies.
Page updated
Google Sites
Report abuse