Academic projects
-
Isometric handgrip exercise speeds working memory responses in younger and older adults.
Decades of animal and pharmacological research have linked the brain’s noradrenergic system to working memory, but there has been relatively little non-pharmacological work in humans examining how isolated fluctuations in noradrenegic activity to working memory performance. We examined how performing isometric handgrip exercise, a manipulation known to influence noradrenergic activity, affected working memory performance. Our results indicated that performing handgrip decreased reaction times on a subsequent working memory task in both younger and older adults. Examining pupil diameter, heart rate and sympathetic tone during and after handgrip, we found a distinct arousal profile that implicated the noradrenergic system in handgrip’s effects on working memory.
publication / preprint / code -
Associations between locus coeruleus MRI contrast and physiological responses to acute stress in younger and older adults.
The noradrenergic locus coeruleus (LC), which serves an arousal hub region in the brain, plays a key role in the response of the central nervous system to acute stress. Yet there has been little work examining how structure of the LC is associated with physiological responses to acute stress. In this project, we examined how multiple measures of physiological arousal during rest, acute stress reactivity and acute stress recovery were related to LC MRI contrast, an in vivo measure of LC structure. Using partial least squares correlation analyses, we found that older participants with higher MRI contrast of the caudal LC had more pronounced physiological responses to acute stress. These findings implicate the LC in neurovisceral integration in aging.
publication / preprint/ code -
Daily heart rate variability biofeedback training decreases locus coeruleus MRI contrast in younger adults.
The brain’s arousal hub region, the locus coeruleus (LC), is bidirectionally connected to the autonomic nervous system, but less in known about factors that affect the LC over time. We tested how a 5-week intervention involving daily heart rate variability biofeedback training affected LC MRI contrast, an in vivo measure of LC structure. Using mixed-effects regression and correlation analyses, we found that LC contrast decreased among participants who performed daily biofeedback, and decreases in LC contrast were related to how much participants increased their heart rate oscillations during training, as well as decreases in sympathetic nervous system signaling. These findings provide novel evidence that LC contrast can be changed over time through an intervention targeting parasympathetic/sympathetic balance.
preprint / code -
Locus coeruleus MRI contrast is associated with cortical thickness in older adults.
Based on growing evidence that the brainstem locus coeruleus (LC) is important for later-life outcomes, we examined relationships between LC MRI contrast, a measure of LC structure, and cortical thickness in younger and older adults from the Berlin Aging Study-II, using multiple linear regression. We found that higher LC contrast was associated with greater cortical thickness in older adults, especially in brain regions that belong to the brain’s frontoparietal attention network.
publication / preprint
Teaching
-
Introductory R & MATLAB tutorial: In 2019, I led a tutorial for research assistants in the Emotion & Cognition Lab focused on the basics of scientific programming in R and MATLAB and creating experiments in Psychtoolbox.
link / github -
Intermediate R tutorial: In 2020, led a tutorial for research assistants in the Emotion & Cognition Lab covering the tidyverse, reproducible reporting with R markdown, and basic statistics
link github
Coding projects in R
-
Word prediction app: As my capstone project for the Data Science Specialization on Coursera, I built a Shiny app that can predict the next words in a sentence, based on a series of inputted words.
app / github -
Predictors of wine quality app: I built a Shiny app to explore and model predictors of Portuguese wine quality, as part of Coursera’s Developing Data Products course.
app / github -
Classification using accelerometer data: As my final project for Coursera’s Practical Machine Learning Course, I used a series of classification models to identify movement types based on accelerometer data.
report -
Tidy Tuesday visualizations: On (some) Tuesdays, I create data visualizations using the tidyverse.
link / github
Coding projects in python
-
Grocery list generator: I wrote a program to create an organized grocery list and prettified image of the list based on user input.
link / demo -
Visualization of Euro-USD exchange rates from 2001-2021: Using data from the European Central Bank, I visualized how exchange rates fluctuated during the terms of the 3 most recent US Presidents.
notebook -
Identifying indicators of heavy traffic on I-94: I identified indicators of heavy traffic volume on US I-94, an east-west interstate highway connecting the Great Lakes and northern Great Plains.
notebook -
Analysis of used car ads on eBay Kleinanzeigen: Using data from the German eBay classifieds site, I analyzed the most common types of advertised used cars and explored price/mileage trends for the top used car brands.
notebook -
Exploration of posts on Hacker News: I used data from the website Hacker News to explore which types of posts received the most comments.
notebook -
Analysis of prison escapes by helicopter: Using data from Wikipedia, I did some basic exploration of worldwide prison escapes that occurred by helicopter.
notebook -
Analysis of mobile app ratings: I used data from the Google Play and Apple App Stores to explore the types of apps that attract the most users.
notebook -
Mapping reward- and loss-related brain activity: As part of Neuromatch Academy’s Computational Neuroscience summer school, I analyzed fMRI data from the Human Connectome Project to examine how monetary reward and loss modulated brain activity.
slides / notebook