/KeepingTrackOfLanguage2023

Data and Code for "Keeping track of language: Can bilingual and monolingual infants associate a speaker with the language they speak?"

Primary LanguageR

Readme for "Keeping track of language: Can monolingual and bilingual infants associate a speaker with the language they speak?"

Created by Esther Schott (esther.schott@mail.concordia.ca)

This repository contains the analysis code associated with the research manuscript Keeping track of language: Can monolingual and bilingual infants associate a speaker with the language they speak?. You can find the associated OSF project, including raw data files and the stimuli used in the studies, at: https://osf.io/psbxv/?view_only=a4fa383dfe7144c7bf0173e1f9c30a43

Repository overview

scripts - this folder contains the scripts used to generate the analysis and manuscript, along with a Groundhog package library for reproducability. To run the analysis, open the KeepingTrackOfLanguage2023.Rproj file located in the root folder using RStudio and then run the lpfp_eyemovement_analysis.Rmd, fp2_pupillometry_analysis.Rmd, lpfp_supplemental.Rmd files to reproduce the analysis.

Pre-processing steps that are not shared

These scripts contain data with identifyable information (date of birth, date of participantion, non-random participant id) and are not shared:

  • lp1_read_eye_data.Rmd reads in eyetracking data, combines files into one (2 projects on tobii, each with different subsets of participants...), exclude attentiongetters and times where no movies is presented
  • lp2_clean_eye_data.Rmd cleaning of participant naming convention, check that no two participants have same name, check that all participants are present
  • lp3_merge_clean_part_list.Rmd data cleaning for demographics, language mixing
  • fp0_merge.Rmd merge eyetracking with info from demographics and master subject list, de-identify data (not shared)

Shared Scripts

  • load_packages.R This is the library of specific packages and versions used in the original analysis script, to make reproducibility of the analysis more straightforward.
  • lp4_participant_exclusion_steps.Rmd read in demographics and main subject list, check for pre-data analysis exclusion criteria (health, language,..)
  • fp1_data_preparation.Rmd merge eye data and participant info, check for pre-data analysis exclusion criteria (health, language,..)
  • lpfp_eyemovement_analysis.Rmd merge lp (Study 1) and fp (Study 2), run analysis on eye movement data
  • fp2_pupillometry_analysis.Rmd pupillometry analysis script
  • lpfp_supplemental.Rmd supplemental materials for manuscript

see below for a schematic overview of the scripts, scripts in blue contain de-identified data and are shared Overview of the order of scripts for repo LPFP

data - this folder contains raw data. however none of these are shared because they contain identifying information. for de-identified datasets for reanalysis, see output/processed data

  • info on trials for Study 2 (FP): facePrince_trial_info.csv
  • age_ranges.csv contains info on age ranges for each age group

output this contains the processed data & output

  • processed_data - This file contains preprocessed data

    • MainSubjList-lp4.Rdata all info on participants for Study 1 (LP)

    • lp_part_info.Rdata de-identified demographics & language mixing info for lp4 script for Study 1 (LP)

    • fp_part_info.Rdata de-identified demographics & language mixing info for fp1 script for Study 2 (FP)

    • fullData-lp2.Rdata eye data for lp5 script

    • fullData_lp5.Rdata output from lp5 script, input for lpfp_eyemovement_analysis.Rmd

    • FacePrince_raw_eye_data.Rdata output from fp0 script, input for fp1_data_preparation.Rmd

    • fp_preprocessed.Rdata output from fp1 script, input for lpfp_eyemovement_analysis.Rmd

    • fp_pupil_preprocessed.Rmd output from fp1 script, input for fp2_pupillometry_analysis.Rmd

    • folder csv_files

  • figures - this folder contains figures generated by the analysis script.

  • tables - this folder contains tables generated by the analysis script.