# Replication Instructions for Hyatt et al. (2022)
This repository contains all code and output for
```
Hyatt, C. S., Listyg, B. S., Owens, M. M., Carter, N. T., Carter, D. R., Lynam, D. R., Harden, K. P., & Miller, J. D. (2022). Structural brain differences do not mediate the relations between sex and personality or psychopathology. Journal of Personality, manuscript in press.
```
Any code-related issues or questions can be directed to Ben Listyg: listyg {at} uga {dot} edu
## Main Analyses
* Download **Main Analysis** folder from https://osf.io/6ftky/
* Upload Human Connectome Project (HCP) Data to the **\Data** folder
* Note: Given HCP file sharing limitations, we are unable to publicly share the full raw data used in our analyses
* CSV files in the **\Data** folder should follow this format:
* `HCP.csv`: Item-level personality, brain morphometry data, and other covariates
* `UPennCNP_HCP_S1200_distribution_ASR_2017_10_5.csv`: Item-level psychopathology data
* In R, set working directory to **\Main Analysis**
* At the console, run `source("Code/run.R")`
## Robustness Checks
* Download **Robustness Checks** folder from https://osf.io/6ftky/
* Upload Human Connectome Data to the **\Data** folder
* Note: Given file sharing constraints with the Human Connectome Project, we are unable to publicly share the full raw data used in our analyses
* CSV files in the **\Data** folder should follow this format:
* `HCP.csv`: Item-level personality, brain morphometry data, and other covariates
* `UPennCNP_HCP_S1200_distribution_ASR_2017_10_5.csv`: Item-level psychopathology data
* For running the analysis that included Glasser Regions, two additional .csv's should be included:
* `glasser_thickness.csv`
* `glasser_area.csv`
* See the **\Glasser Regions Instruction** folder for more information
* For additional information on the Glasser Regions, please contact Dr. Max Owens
* In R, set working directory to **\Robustness Checks\{Desired Sub-folder}**
* At the console, run `source("Code/run.R")`
* Note: Given the large number of mediation models being fit, computation time can be quite long (especially in the case of the Glasser Region analysis). Please prepare accordingly.
