DDM: Individual subject Drift Diffusion Model (DDM)
In jeff324/powder: Power posterior sampling via differential evolution

Description Usage Arguments Format Fields Methods Examples

Individual subject Drift Diffusion Model (DDM)

DDM

`a`	threshold seperation
`v`	drift rate
`t0`	non-decision time
`z`	starting point
`st0`	inter-trial-variability of non-decision time
`sv`	inter-trial-variability of drift rate
`sz`	inter-trial-variability of starting point
`prior`	list containing priors

R6Class object.

theta.names: a character vector containing the names of the subject-level parameters
theta.init: a function that provides a random initial value for each subject-level parameter
theta.start.point: a numeric vector containing means of start points used to initialize in theta.init
vary.parameter: a logical vector containing parameters to vary
prior: a list containing priors on all parameters

log.dens.prior(x,hyper): likelihood of subject-level parameters given group-level parameters
log.dens.like(x,data,par.names): LBA likelihood function

## Not run: 
# DDM model that varies threshold and drift rate across 3 conditions
model = DDM.Individual$new(a=T,v=T,conds=1:3)

# Note, inter-trial-variability components, st0, sv, and sz
# have 3 options: TRUE, FALSE, or 0. When TRUE,
# they will vary across conditions. When FALSE
# they are fit, but not allowed to vary across conditions.
By default, they are set to 0.

# st0 is a parameter, other variability parameters are 0
model_st0 = DDM$new(st0=FALSE)

# sv is allowed to vary over conditions
model_sv = DDM$new(sv=TRUE)

# Setting priors:
prior = list(
    a = list(mu = c(mu = 1, sigma = 1),
             sigma = c(mu = 1, sigma = 1)),
    v = list(mu = c(mu = 1, sigma = 1),
             sigma = c(mu = 1, sigma = 1)),
    t0 = list(
         mu = c(mu = .3, sigma = .3),
         sigma = c(mu = .3, sigma = .3)),
    st0 = list(
         mu = c(mu = .1, sigma = .1),
         sigma = c(mu = .1, sigma = .1)))

 # a and v vary across conditions
 # sz and sv are 0
 # t0 and st0 are parameters that do not vary across conditions
 model = DDM$new(a=TRUE,v=TRUE,prior=prior)

 # Priors can also be changed after creating the model
 model$prior$a$mu = c(mu=2,sigma=2)


## End(Not run)