# Workflow for "Distinct Monitoring Strategies underlie costs and performance in prospective memory"
## Analysis Steps
### Step 1: Data Summary Creation (in Python)
1. The first step is to unzip the data files in the eyeData, behavioralData, and summaryData folders.
2. To perform analyses from scratch, you can first preprocess the data from the raw experimental matlab files to the summary data that is used for the analyses. I performed analyses in Python v3. You shouldn't need to install many new packages, I believe you'll just need: numpy, os, re, pandas, glob, scipy, and tqdm). Or you can do your own thing! Initial preprocessing done for the submitted manuscript can be replicated by first generating the by-probe and by-trial summaries.
- Script: ~/Scripts/preprocessingData.py
- Output 1: ~/MemorEyes/Data/summaryData/byProbe.csv
- Output 2: ~/MemorEyes/Data/summaryData/byTrial.csv
2. The next step is to generate a separate dataframe/file that has the fixation location by time for each probe. This is separate to keep the previous dataframes from becoming too unwieldy.
- Script: ~/Scripts/preproc_fixLocXtime.py
- Output: ~/Data/summaryData/fixLocXtime.csv
### Step 2: Data Analysis (in R)
- I apologize, because at the moment everything is just bundled together in a long R-script (~/MemorEyes/Scripts/AnalysisCode.R). For the most part, analyses appear in the approximate order that they are reported in the manuscript. To replicate analyses reported in the manuscript, you should be able to just run the script.
- The bootstrap analyses can take a while, especially the big one comparing fixations/pmCosts to PM accuracy. I've included the bootstrap results that were generated from my analyses in the 'MemorEyes/Data/summaryData' folder and indicated where they go in the script. Alternatively, I've included the bootstrap code that I used so you can replicate/examine it if you want to.
