Fall 2015 group project
Team Member: Jon Jara, Juan Shishido, Wendy Xu
In the data/
directory, you can make data
to download and extract the data
files and make validate
to chech hashes.
In the code/
directory, there are several options:
$ make behavioral_regression
to run logstic regression on the behavorial
data for each subject$ make pda_outliers
for plotting outlier volumes for 2 subjects$ make pda_smooth
to plot a brain image before and after smoothing$ make neural_regression
to run the regression for all subjects, returning
a gridplot for the beta coefficients$ make mvpa
for our (experimental) MVPA analysis and plots on all 16
subjectsIn the report/
directory, make
to create the report PDF.
Our group is working on the paper The Neural Basis of Loss Aversion in Decision-Making Under Risk, which investigates whether loss aversion reflects the engagement of distinct emotional processes when potential losses are considered and examines the neural systems that process decision utility. The paper can be found in our project
The study has 16 subjects (eight males and eight females with an average age of 22). For each of them, three trials of the mixed gambles task are performed and data sets of blood-oxygen-level-dependent are collected and contrasted. The study uses two modeling approaches for the primary whole-brain analyses---the parametric analysis and the matrix analysis. The paper can be found in Project-eta.
We plan to reproduce this study based on the whole-brain statistical analysis. We will look into the two modeling approaches that the authors used, and come up with our own approach based on what we have and what we can do.
Here are a few things we are doing:
Data cleaning, normalization
GLM
Correlate behavioral risk aversion vs neural risk aversion
Multi-voxel pattern analysis