In kleinbub/rMEA: Synchrony in Motion Energy Analysis (MEA) Time-Series

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "tools/readme/README-"
)

rMEA

The goal of rMEA is to provide a suite of tools useful to read, visualize and export bivariate Motion Energy time-series. Lagged synchrony between subjects can be analyzed through windowed cross-correlation. Surrogate data generation allows an estimation of pseudosynchrony that helps to estimate the effect size of the observed synchronization.

Example

This example shows a complete analysis pipeline consisting on Motion Energy time-series import, pre-processing, cross-correlation analysis and comparison between groups against pseudosynchrony.

library(rMEA)

## read the first sample (intake interviews of patients that carried on therapy)
path_normal <- system.file("extdata/normal", package = "rMEA")
mea_normal <- readMEA(path_normal, sampRate = 25, s1Col = 1, s2Col = 2,
                     s1Name = "Patient", s2Name = "Therapist", skip=1,
                     idOrder = c("id","session"), idSep="_")
mea_normal <- setGroup(mea_normal, "normal")

## read the second sample (intake interviews of patients that dropped out)
path_dropout <- system.file("extdata/dropout", package = "rMEA")
mea_dropout <- readMEA(path_dropout, sampRate = 25, s1Col = 1, s2Col = 2,
                     s1Name = "Patient", s2Name = "Therapist", skip=1,
                     idOrder = c("id","session"), idSep="_")
mea_dropout <- setGroup(mea_dropout, "dropout")

## Combine into a single object
mea_all <- c(mea_normal, mea_dropout)

summary(mea_all)

## Show diagnostics for the first session:
diagnosticPlot(mea_all[[1]])
plot(mea_all[[1]], from=1, to=200)

## Filter the data
mea_smoothed <- MEAsmooth(mea_all)
mea_rescaled <- MEAscale(mea_smoothed)

## Generate a random sample
mea_random <- shuffle(mea_rescaled, 50)

## Run CCF analysis
mea_ccf <- MEAccf(mea_rescaled, lagSec= 5, winSec = 30, incSec=10, ABS = F)
mea_random_ccf <- MEAccf(mea_random, lagSec= 5, winSec = 30, incSec=10, ABS = F)

## Visualize results

# Raw data of the first session with running lag-0 ccf
plot(mea_ccf[[1]], from=100, to=300, ccf = "lag_zero")

# Heatmap of the first session
MEAheatmap(mea_ccf[[1]])

# Distribution of the ccf calculations by group, against random matched dyads
MEAdistplot(mea_ccf, contrast = mea_random_ccf)

# Representation of the average cross-correlations by lag
MEAlagplot(mea_ccf, contrast=mea_random_ccf)