Skip to content

Latest commit

 

History

History
42 lines (28 loc) · 1.76 KB

README.md

File metadata and controls

42 lines (28 loc) · 1.76 KB

A toolkit for nonlinear analyses

This repository contains the necessary code and tools to begin using categorical recurrence quantification analysis and fractal analysis (detrended fluctuation analysis).

Accompanying manuscript: Chiovaro, M., & Paxton, A. (under review). Nonlinear, Natural, and Noisy: A Quantitative Approach to the Collection and Analysis of Real-World Social Behavior.

Presented at: The 49th Annual Meeting of the Society for Computers in Psychology (SCiP) (Montréal, Québec, Canada)

Using this repository

To get started using this repository, clone it and create the following subdirectories:

  • ./data/raw/
  • ./data/formatted/
  • ./results/rqa/
  • ./results/dfa/

Data preparation

These analyses start with time-series of sound on-off markers in .txt format (generated using Audacity)

Load your sound-marker data into: ./data/raw/

To get your data into the appropriate format, follow:

  • ./scripts/01-formatting-data/formatting_rqa.R: for categorical recurrence quantification analysis.
  • ./scripts/01-formatting-data/formatting_dfa.R: for generating "Inter-Onset Interval" and "Turn Duration" on full time-series and truncated (25k samples) data for fractal analysis (detrended fluctuation analysis, DFA). These formatted data will be saved to ./data/formatted/

Analyses

To run the analyses, adapt:

  • ./scripts/02-analyses/catRQA/analysis.R
  • ./scripts/02-analyses/fractal_analysis/analysis.R

To compare fractal results with results of categorical RQA, we ran DFA on the truncated 25k sample using:

  • ./scripts/02-analyses/fractal_analysis/analysis_truncated.R

All formatted data/images/results will be saved to:

  • ./results/rqa
  • ./results/dfa

Open science = open conversation!

Feel free to submit a pull request to contribute!