SimMLDS function

Simulate Output of MLDS Experiment

Given a block of trials of an MLDS experiment, an underlying response function and the judgment variability, simulate the response of an observer.

SimMLDS(Trials, Scale, Sigma, n = 1)

Arguments

• Trials: an N by 4 or 3 matrix or data frame of integers indicating the n trials of an MLDS experiment. The columns indicate the indices of the stimuli presented on a trial, 4 for an experiment with quadruples and 3 for triads. A data frame for this argument is most easily generated with the combn function.
• Scale: a vector of values indicating the underlying responses of the simulated observer for each stimulus level. The length of this vector should equal the largest integer in Trials.
• Sigma: a vector of length 1 indicating the judgment standard deviation of the simulated observer.
• n: integer giving number of simulated data sets to return

Details

Given a data frame of indices to the responses associated with stimulus levels and the judgment variability, the function returns the results of 1 or multiple MLDS experiments, either with triads or quads, depending on the number of columns in the data frame.

Returns

If the argument n is set to 1 (default), an object of class mlds.df or mlbs.df with simulated responses. If n is greater than 1, a list of such objects is returned.

References

Maloney, L. T. and Yang, J. N. (2003). Maximum likelihood difference scaling. Journal of Vision, 3(8):5 , 573--585, tools:::Rd_expr_doi("10.1167/3.8.5") .

Knoblauch, K. and Maloney, L. T. (2008) MLDS: Maximum likelihood difference scaling in R. Journal of Statistical Software, 25:2 , 1--26, tools:::Rd_expr_doi("10.18637/jss.v025.i02") .

Author(s)

Kenneth Knoblauch and Laurence T. Maloney

See Also

see also boot.mlds

Examples

Tr <- t(combn(10, 4))
Sc <- seq(0, 1, len = 11)^2
Sig <- 0.2
sim.lst <- SimMLDS(Tr, Sc, Sig, n = 10)
sim.res <- sapply(sim.lst, mlds)