View article

A straightforward extension of the signal detection theory (SDT) framework is described and demonstrated for the two-alternative identification task. The extended framework assumes that the subject and an arbitrary model (or two subjects, or the same subject on two occasions) are performing the same task with the same stimuli, and that on each trial they both compute (in effect) values of a decision variable. Thus, their joint performance is described by six fundamental quantities: two levels of intrinsic discriminability (" d-prime"), two values of decision criterion, and two decision-variable correlations, one for each of the two categories of stimuli in the task. Decision-variable correlations (DVCs) provide increased power for testing models and for characterizing individual differences, and do not require a special experimental design (eg, they can be computed from existing data). The extended framework was developed …