WPT {hmmr}

Weather Prediction Task Data

Description

This data set contains responses of 11 Parkinsons' patients and 13 age-matched controls on the Weather Prediction Task. Both groups were tested twice. The PD patients were either on or off dopaminergic medication.

Usage

data(WPT)

Format

A data.frame with 9600 observations on the following variables.

id

a factor with participant IDs

group

a factor with group IDs (Parksinson's patient or control)

med

a factor indicating, for the PD patients, whether they were on dopaminergic medicine or not

occ

a numeric vector with testing occassions

trial

a numeric vector with trial numbers

c1

a numeric (binary) vector indicating whether the first cue was present (1) or not (0)

c2

a numeric (binary) vector indicating whether the second cue was present (1) or not (0)

c3

a numeric (binary) vector indicating whether the third cue was present (1) or not (0)

c4

a numeric (binary) vector indicating whether the fourth cue was present (1) or not (0)

y

a factor with the actual outcome (Rainy or Fine)

r

a factor with participants' prediction of the outcome

Source

Speekenbrink, M., Lagnado, D. A., Wilkinson, L., Jahanshahi, M., & Shanks, D. R. (2010). Models of probabilistic category learning in Parkinson's disease: Strategy use and the effects of L-dopa. Journal of Mathematical Psychology, 54, 123-136.

Corresponding author: [e-mail]

Examples

data(WPT)

# set up predictors for the different strategies
WPT$sngl <- 0 # singleton strategy
WPT$sngl[WPT$c1 == 1 & rowSums(WPT[,c("c1","c2","c3","c4")]) == 1] <- -1
WPT$sngl[WPT$c2 == 1 & rowSums(WPT[,c("c1","c2","c3","c4")]) == 1] <- -1
WPT$sngl[WPT$c3 == 1 & rowSums(WPT[,c("c1","c2","c3","c4")]) == 1] <- 1
WPT$sngl[WPT$c4 == 1 & rowSums(WPT[,c("c1","c2","c3","c4")]) == 1] <- 1
WPT$sc1 <- 1 - 2*WPT$c1
WPT$sc2 <- 1 - 2*WPT$c2
WPT$sc3 <- -1 + 2*WPT$c3
WPT$sc4 <- -1 + 2*WPT$c4
WPT$mc <- sign(-WPT$c1 - WPT$c2 + WPT$c3 + WPT$c4)

rModels <- list(
	list(GLMresponse(formula=r~-1,data=WPT,family=binomial())),
	list(GLMresponse(formula=r~sngl-1,data=WPT,family=binomial())),
	list(GLMresponse(formula=r~sc1-1,data=WPT,family=binomial())),
	list(GLMresponse(formula=r~sc2-1,data=WPT,family=binomial())),
	list(GLMresponse(formula=r~sc3-1,data=WPT,family=binomial())),
	list(GLMresponse(formula=r~sc4-1,data=WPT,family=binomial())),
	list(GLMresponse(formula=r~mc-1,data=WPT,family=binomial()))
)

transition <- list()
for(i in 1:7) {
  transition[[i]] <- transInit(~1,nstates=7,family=multinomial(link="identity"))
}

inMod <- transInit(~1,ns=7,data=data.frame(rep(1,48)),family=multinomial("identity"))

mod <- makeDepmix(response=rModels,transition=transition, 
	prior=inMod,ntimes=rep(200,48),stationary=TRUE)

fmod <- fit(mod)