PlotSensitivityDecline: Show decline in visual sensitivity from choice history biases

In armanabraham/chb: Choice history biases

Description

Displays decline in visual sensitivity as a result of choice history biases. It can show reduction in contrast sensitivity (75 change in slope of psychometric function (in case of probit, smaller slope means less sensitivity).

Usage

PlotSensitivityDecline(simThAndSlope, whatToPlot = "slope",
  whatToReturn = "plot")

Arguments

simThAndSlope	simulated threshold and slope computed using function NoBiasVsSubjectBias (see example below).
whatToPlot	either decline of slope or threshold can be plotted
whatToReturn	return either 'plot' as ggplot object or 'data' prepared for plotting, which can be used for other operations, or 'stats'. 'stats' parameter shows p value of one-sampled test of median change (whether sensitivity decline is significantly different from 0)

Examples

load('20150919_allWeights_RIKEN_UCL_Stanford_cond1n13.RData',verbose=TRUE)
regWeights <- droplevels(subset(allWeights, Regularized=='yes'))
oneSbjWeights <- droplevels(subset(regWeights, SubjectID %in% c('s001', 's002', 's003', 's009', 's008', 's007')))
thAndSlope <- NoBiasVsSubjectBias(regWeights, nTrialsPerContrast=10, B=5) # B=5 is for fast simulation, for meaningful results use B>500.
PlotSensitivityDecline(thAndSlope) #

armanabraham/chb documentation built on May 10, 2019, 1:39 p.m.