ssp_dm: Create the Shrinking Spotlight Model

In dRiftDM: Estimating (Time-Dependent) Drift Diffusion Models

Create the Shrinking Spotlight Model

Description

This function creates a drift_dm object that corresponds to a simple version of the shrinking spotlight model by \insertCiteWhiteetal.2011;textualdRiftDM.

Usage

ssp_dm(
  var_non_dec = TRUE,
  var_start = FALSE,
  instr = NULL,
  obs_data = NULL,
  sigma = 1,
  t_max = 3,
  dt = 0.005,
  dx = 0.02,
  b_coding = NULL
)

Arguments

var_non_dec, var_start	logical, indicating whether the model should have a variable non-decision time or starting point (see also nt_uniform and x_uniform in component_shelf
instr	optional string with "instructions", see modify_flex_prms().
obs_data	data.frame, an optional data.frame with the observed data. See obs_data.
sigma, t_max, dt, dx	numeric, providing the settings for the diffusion constant and discretization (see drift_dm)
b_coding	list, an optional list with the boundary encoding (see b_coding)

Details

The shrinking spotlight model is a model developed for the flanker task.

It has the following properties (see component_shelf):

• a constant boundary (parameter b)

• a constant starting point in between the decision boundaries

• an evidence accumulation process that is driven by an attention spotlight that covers both the flankers and the target. The area that covers the flankers and target is modeled by normal distribution with mean 0:

  • At the beginning of the trial attention is wide-spread, and the width at t=0 is the standard deviation sd_0

  • As the trial progresses in time, the attention spotlight narrows, reflected by a linear decline of the standard deviation with rate r (to a minimum of 0.001).

  • the attention attributed to both the flankers and the target is scaled by p which controls the strength of evidence accumulation

• A non-decision time that follows a truncated normal distribution with mean non_dec and standard deviation sd_non_dec.

• The model also contains the auxiliary parameter sign, which is used to control the influence of the flankers across conditions. It is not really a parameter and should not be estimated!

Per default, the parameter r is assumed to be fixed (i.e., is not estimated freely). The model also contains the custom parameter interf_t, quantifying the interference time (sd_0 / r).

Value

An object of type drift_dm (parent class) and ssp_dm (child class), created by the function drift_dm().

Note

The parameters of SSP in dRiftDM differ in their size from the original publication of \insertCiteWhiteetal.2011;textualdRiftDM. dRiftDM uses symmetrical boundaries around zero and a diffusion constant of 1. In the original publication, SSP was parameterized with boundaries ranging from zero to a and a diffusion constant of 0.1.

Thus, in dRiftDM, the boundary b corresponds to b = a/2 \cdot 10. Additionally, p in dRiftDM is 10 times larger than p in the original publication. Finally, r is expressed in seconds, and thus r is 1000 times larger in dRiftDM than in the original publication.

References

\insertRef

Whiteetal.2011dRiftDM

Examples

# the model with default settings
my_model <- ssp_dm()

# the model with a finer discretization
my_model <- ssp_dm(dt = .0025, dx = .01)

dRiftDM documentation built on Dec. 1, 2025, 5:08 p.m.