VWPre: VWPre: Tools for Preprocessing Visual World Data.

In VWPre: Tools for Preprocessing Visual World Data

Description

The VWPre package provides a set of functions for preparing Visual World data collected with SR Research Eyelink eye trackers.

Formatting functions

• The function create_time_series returns a time columns in milliseconds.

• The function prep_data returns a data table with correctly assigned classes for important columns.

• The function relabel_na returns a data table with samples containing 'NA' relabeled as outside any interest area.

• The function recode_ia returns a data table containing recoded interest area IDs and/or interest area labels.

• The function select_recorded_eye returns a data table with data from the the recorded eye in new columns (IA_ID and IA_LABEL).

• The function custom_ia returns a data table with gaze data remapped to new interest areas.

• The function align_msg returns a data table with newly aligned sample data in a new column (Align).

• The function rm_extra_DVcols removed DataViewer coumns that are not necessary for preprocessing with this package.

Calculation functions

• The function bin_prop returns a downsampled data table containing proportion of looks (samples) to each interest area in a particular window of time (bin size).

• The function transform_to_elogit returns a data table with proportion looks transformed to empirical logits with weights.

• The function create_binomial returns a data table with a new success/failure column for each IA which is suitable for logistic regression.

Trackloss functions

• The function mark_trackloss returns a data table with data information regarding trackloss of the sample.

• The function rm_trackloss_events returns a data table from which events without the minimum amount of quality data have been removed.

Fasttrack formatting function

• The function fasttrack a meta-function that returns a data table of processed data containing the result of the series of necessary subroutines. This is intended for experienced users doing basic preprocessing.

Data-checking functions

• The function check_eye_recording returns a summary of whether or not the dataset contains gaze data in both the Right and Left interest area columns.

• The function check_time_series returns the first value in the Time column for each event.

• The function check_samples_per_bin returns the number of samples in each bin.

• The function check_samplingrate returns the value corresponding to the sampling rate in the data.

• The function ds_options returns the binning (downsampling) options possible for the given sampling rate.

• The function check_ia returns a summary of the interest area IDs and Labels present in the data.

• The function check_msg_time returns a summary of the the time value at a given sample message for each recording event.

• The function check_all_msgs returns all messages in the data and their time stamp.

Plotting functions

• The function plot_avg returns a plot of the grand or conditional averages of proportion (or empirical logit) looks to each interest area along with error bars.

• The function plot_avg_contour returns a contour plot of the conditional average of proportion (or empirical logit) looks to a given interest area over Time and a specified continuous variable.

• The function plot_avg_diff returns a plot of the grand or conditional averages of the difference between looks to two interest areas (proportions or empirical logits) with error bars.

• The function plot_avg_cdiff returns a plot of the average difference between two conditions for looks to a given interest area (proportions or empirical logits) with error bars.

• The function make_pelogit_fnc returns a function that can backtransform predicted empirical logit to probability scale, particularly (though not exclusively) useful for plotting purposes.

Interactive functions

• The function plot_transformation_app opens a Shiny app for visualizing the effect of both number of observations and constant on the results of the empirical logit transformation and weight calculations.

• The function plot_indiv_app opens a Shiny app for inspecting by-subject or by-item averages for all interest areas, alongside the grand average (for proportion or empirical logit looks) within a specified time window.

• The function plot_var_app opens a Shiny app for inspecting by-subject or by-item Z-scores with respect to the overall mean for a given interest area within a specified time window.

Notes

• The vignettes are available via browseVignettes(package = "VWPre").

• A list of all available functions is provided in help(package = "VWPre").

• This package can be cited using the information obtained from citation("VWPre") or print(citation("VWPre"), bibtex = TRUE)

Author(s)

Vincent Porretta, Aki-Juhani Kyröläinen, Jacolien van Rij, Juhani Järvikivi

Maintainer: Vincent Porretta ([e-mail])

University of Windsor, Canada

VWPre documentation built on Nov. 30, 2020, 1:08 a.m.