dprime.SD {psyphy}

d' for Same-different Paradigm

Description

Calulate d' for same-different paradigm either assuming a differencing strategy or independent observations

Usage

dprime.SD(H, FA, zdiff, Pcmax, method = "diff")

Arguments

H

numeric in [0, 1] corresponding to Hit rate

FA

numeric in [0, 1] corresponding to False alarm rate

zdiff

numeric. Difference of z-scores for Hit and False Alarm rates ( only valid for method "IO")

Pcmax

numeric in [0, 1]. Proportion correct for an unbiased observer, pnorm(zdiff/2) (only valid for method "IO").

method

character. Specifies the model to describe the observer's criterion for dividing up the decision space, must be either "diff" for a differencing strategy (the default) or "IO" for independent observations.

Details

Two different strategies have been described for how the observer partitions the decision space in the same-different paradigm. With Independent Observations, d' can be calculated either from the H and FA rates, from the difference of z-scores or from the probability correct of an unbiased observer. Only one of these three choices should be specified in the arguments. For the differencing strategy, only the first of these choices is valid.

Value

Returns the value of d'

Author(s)

Kenneth Knoblauch

References

MacMillan, N. A. and Creeman, C. D. (1991) Detection Theory: A User's Guide Cambridge University Press

Green, D. M. and Swets, J. A. (1966) Signal Detection Theory and Psychophysics Robert E. Krieger Publishing Company

See Also

dprime.mAFC, dprime.ABX, dprime.oddity

Examples

dprime.SD(H = 0.642, F = 0.3)
dprime.SD(H = 0.75, F = 0.3, method = "IO")
dprime.SD(zdiff = qnorm(0.75) - qnorm(0.3), method = "IO")
dprime.SD(Pcmax = pnorm( (qnorm(0.75) - qnorm(0.3))/2 ), method = "IO")