Skip to contents

Simulated response times from four participants, with ground truth

Source: R/data.R
rt_example.Rd

A small data set for the examples and the vignette: four participants, two conditions, 100 trials per cell, generated by r_contaminated() from a diffusion core with contaminants from all three processes. Because the data are simulated, every trial carries the truth a real data set cannot: whether it was a contaminant, and which process produced it.

Usage

rt_example

Format

A data.frame with 800 rows and 9 columns:

id

participant, "p1" to "p4".

condition

factor, "hard" or "easy"; the base drifts the two conditions are built from differ by 0.4, before the participant's ability factor multiplies both (see Details).

trial

trial number within the cell, 1 to 100.

rt

response time in seconds.

response

1 for a correct response, 0 for an error.

contaminant

logical ground truth.

process

"clean", "leading_edge", "delay", or "informationless".

true_drift

the drift the cell was generated from.

contam_rate

the participant's contamination rate: .02, .05, .10, and .15 for p1 to p4.

Details

The participants differ in two things at once. Their drift differs, by a fixed ability factor of 0.85, 0.95, 1.05, and 1.20 on a base drift of 1.5 (hard) and 1.9 (easy), with bound = 1.2 and ndt = 0.30 throughout. And their contamination rate differs, so that the error in a participant's estimate can be set against their own rate, as the get-started vignette does.

The generating script is data-raw/rt_example.R in the source repository; the seed is fixed there, not in the package.

See also

r_contaminated() to generate data shaped like your own task.

Examples

head(rt_example)
#>   id condition trial    rt response contaminant process true_drift contam_rate
#> 1 p1      easy     1 0.590        1       FALSE   clean      1.615        0.02
#> 2 p1      easy     2 0.408        1       FALSE   clean      1.615        0.02
#> 3 p1      easy     3 1.334        1       FALSE   clean      1.615        0.02
#> 4 p1      easy     4 0.563        1       FALSE   clean      1.615        0.02
#> 5 p1      easy     5 0.574        0       FALSE   clean      1.615        0.02
#> 6 p1      easy     6 0.665        1       FALSE   clean      1.615        0.02
table(rt_example$id, rt_example$process)
#>     
#>      clean delay informationless leading_edge
#>   p1   194     2               4            0
#>   p2   190     3               2            5
#>   p3   182     7               5            6
#>   p4   174     7               9           10