Gardner, Dalsing, Reyes, and Brake (1984) where Hand F are the ordinates of the standard normal distribution at the points where the tail areas of the distribution are equivalent to the hit rate and the false-alarm rate, respectively. That is, H is the ordinate of the signalplus-noise distribution at the criterion, and F is the ordinate of the noise-alone distribution at the criterion. Gescheider (1985, chap. 4) and Hochhaus (1972) concisely described computation of {3.To test the accuracy of the Gardner et al. (1984) The magnitude of {3 indicates the stringency of the criterion an observer uses in deciding to report detection of a signal. High {3 signifies a strict criterion; low {3 signifies a lax criterion. {3 can be determined from the relationship between (1) the proportion of trials in which the observer reports a signal when the signal is present (the hit rate) and (2) the proportion of trials in which the observer reports a signal when the signal is absent (the falsealarm rate). Values of {3 are calculated by the equationrors occur primarily at low levels of the false-alarm rate, but every page of the table contains some errors.
ALTERNATIVES TO THE TABLEAlthough the errors in the Gardner et al, (1984)