The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.

dRiftDM

The package dRiftDM was developed to assist psychological researchers in applying and fitting diffusion models to empirical data within the R environment. Its most important feature is the ability to handle non-stationary problems, specifically diffusion models with time-dependent parameters. The package includes essential tools for standard analyses, such as building models, estimating parameters for multiple participants (individually for each participant), and creating summary statistics. The pre-built models available in the package include:

• The Standard (Ratcliff) Diffusion Model (Ratcliff, 1978, Psychological Review)
• The Diffusion Model for Conflict Tasks (Ulrich et al., 2015, Cognitive Psychology)
• The Shrinking Spotlight Model (White et al., 2011, Cognitive Psychology)

Users can flexibly create custom models and utilize the dRiftDM machinery for estimating them.

Starting with version 0.2.0, model predictions (i.e., first-passage times) are derived by numerically solving the Kolmogorov-Forward Equation or a coupled set of integral equations, based on code provided by Richter et al. (2023, Journal of Mathematical Psychology).

Notes

Compared to the previous version 0.1.1, versions >0.2.0 make greater use of the S3 object system. Additionally, beginning with version 0.2.0, models use “flex_prms” objects to handle parameters across conditions.

If you wish to install the older version (0.1.1), you can use:

devtools::install_github("bucky2177/dRiftDM", ref = "0.1.1")

Installation

You can install the development version of dRiftDM from GitHub with:

# install.packages("devtools")
devtools::install_github("bucky2177/dRiftDM")

The CRAN version can be installed with:

install.packages("dRiftDM")

Example

We are in the process of publishing a tutorial with more detailed instructions (see the respective OSF pre-print). For now, here is an example of how to fit the Diffusion Model for Conflict Tasks to some synthetic data using bounded Nelder-Mead:

library(dRiftDM)
#> 
#>  ---------------------------------------------------- 
#> / Welcome to dRiftDM 0.2.1 Please report any bugs or \
#> \ unexpected behavior                                /
#>  ---------------------------------------------------- 
#>       \
#>        \
#> 
#>         ^__^ 
#>         (oo)\ ________ 
#>         (__)\         )\ /\ 
#>              ||------w|
#>              ||      ||
dmc_model <- dmc_dm(dx = .002, dt = .002, t_max = 1.2)
obs_data(dmc_model) <- dmc_synth_data
print(dmc_model)
#> Class(es): dmc_dm, drift_dm
#> 
#> Current Parameter Matrix:
#>        muc   b non_dec sd_non_dec  tau a    A alpha
#> comp     4 0.6     0.3       0.02 0.04 2  0.1     4
#> incomp   4 0.6     0.3       0.02 0.04 2 -0.1     4
#> 
#> Unique Parameters:
#>        muc b non_dec sd_non_dec tau a A alpha
#> comp   1   2 3       4          5   0 6 7    
#> incomp 1   2 3       4          5   0 d 7    
#> 
#> Special Dependencies:
#> A ~ incomp == -(A ~ comp)
#> 
#> Custom Parameters:
#>        peak_l
#> comp     0.04
#> incomp   0.04
#> 
#> Deriving PDFs:
#>   solver: kfe
#>   values: sigma=1, t_max=1.2, dt=0.002, dx=0.002, nt=600, nx=1000
#> 
#> Observed Data: 300 trials comp; 300 trials incomp

est_model <- estimate_model(
  drift_dm_obj = dmc_model, # a model
  lower = c(2, 0.4, 0.2, 0.005, 0.02, 0.02, 2), # lower boundary - search space
  upper = c(6, 0.8, 0.5, 0.050, 0.12, 0.30, 7), # upper boundary - search space
  verbose = 2, # print out each evaluation
  use_de_optim = F, use_nmkb = T # use Nelder-Mead for demonstration
)

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.