# Zoo Free Recall Analysis scripts for the main manuscript results section and figures. (Note: Values related to permutation scripts reported in this component will vary from the manuscript, as noted below. Please see component 1 for values used in manuscript analysis) ## Setup 1. Download the entire Rproject folder from the OSF at [link](link) 2. Install necessary versions of r packages (manually or via `renv::restore()`) ## Install Packages This project uses the *renv* package to manage package versions. All necessary information regarding packages used anywhere within this project are saved in the `renv.lock` file. For more information about the *renv* package, check out <https://rstudio.github.io/renv/articles/renv.html>. To install all of the proper package versions, run the following code. **Be sure to have the Rproject open in RStudio prior to running this code**. It will automatically download and install the needed R packages. It will also functionally separate this R project from your other projects (you won’t have access to other packages already installed on your machine while still in this project). ``` r renv::restore() ``` If using *renv* is not desirable, you can also manually install the necessary packages using the following code: ``` r install.packages(c("tidyverse","DT","here","psych","rstatix","jtools","xlsx")) ``` ## Usage Once the project has been downloaded and the r packages installed, running the following code can be used to recreate the cleaned data and the results section of the manuscript: **Note** - Running the cleaning script can take a little while (about 10-15 minutes). Since it randomly permutes the subjects’ responses, there will be small numerical discrepancies between any analyses using the permuted distributions and the published manuscript. The pattern of results and statistical significance of all results should remain consistent. There is a `set.seed(7272)` function used to ensure computational reproducibility, but that does not always ensure exactly identical results. ``` r # load the data and clean it source(here("scripts","clean_and_save_data.R")) # rerun the results script and save it into the reports folder rmarkdown::render( here::here("script files", "manuscript_results.Rmd"), output_file = "manuscript_results", output_dir = here::here("reports"), ) ``` ## Understanding the Layout of this Project This R-project has four main subdirectories: `/data`, `/reports`, `/figures`, and `/script files`. ### `/data` - All subject IDs for the data in this archive have been randomized to add a further layer of anonymity - `/data_manuscript` - `data_manuscript_no-mulitple-locations.xlsx` - This is the full data. - `data_manuscript_no-mulitple-locations_cleaned.rds` is the result of running the `clean_and_save_data.R` script. It will retain the proper data formats and structures for all analyses and graphs which use this data. ### `/reports` - This folder holds all reports which were generated within this project. The only report included is the one created by the `manuscript_results.Rmd` script. - `manuscript_results.html` - This HTML file contains the code and results from the results section of the published manuscript. The serial position analyses and the post-hoc partial correlation analyses are not included but can be found at with the osf project at [LINK](!link) and [LINK](!link) ### `script files` - `clean_and_save.R` - This script scrapes the `/data` directory (and it’s subdirectories) for all `.xlsx` files, imports them into R, cleans them, generates the permutations (for context, forward, and backward adjacency scores), computes the absolute lag differences, and percent rank of the first recall location. This script all saves each cleaned result as an `.rds` file in the same directory as the original .xlsx file with a suffix of “\_cleaned” appended just before the file extension. - `generate_manuscript_figure_1.R` - This script recreates the following `\figures` folder: - `figure_1_300dpi.jpg` - Figure 1 that appears in manuscript - `figure_1_4-5_300dpi.jpg` - faceted version of Figure 1 highlighting 4-5-year-old permutation - `figure_1_6-7_300dpi.jpg` - faceted version of Figure 1 highlighting 6-7-year-old permutation - `figure_1_8-10_300dpi.jpg` - faceted version of Figure 1 highlighting 8-10-year-old permutation - `figure_2_alternate_300ppi.jpg` - alternate (unused) version of figure 2 - `generate_manuscript_figure_2.R` - This script recreates the following `\figures` folder: - `figure_2a_error_bars_300dpi.jpg` - `figure_2a_without_error_bars_300dpi.jpg` - `figure_2b_error_bars_300dpi.jpg` - `figure_2b_without_error_bars_300dpi.jpg` - `helper_functions.R` - This file holds all of the functions used by this project (e.g., functions used to create permutations, or compute context scores, or generate reports) - `manuscript_results.Rmd` - This script runs the main analyses within the manuscript. The output is an HTML file which can be viewed in any web browser. It has a clickable table of contents to bring users quickly to particular sections of the results. There are also buttons which can show/hide the code used for each section of the results. - `randomize_subject_ids.R` - This script checks the project for any `.xlsx` files and randomizes the subject IDs within each age group. This script is **NOT** intended to be run by users. It was included to show the process used to randomize subject IDS prior to the data being posted online. ## Acknowledgments - README.md template from <https://gist.github.com/PurpleBooth/109311bb0361f32d87a2>