Signal Detection Theory Analysis of Type 1 and Type 2 Data: Meta-d′, Response-Specific Meta-d′, and the Unequal Variance SDT Model

摘要: Previously we have proposed a signal detection theory (SDT) methodology for measuring metacognitive sensitivity (Maniscalco and Lau, Conscious Cogn 21:422–430, 2012). Our SDT measure, meta-d′, provides response-bias free measure of how well confidence ratings track task accuracy. Here provide an overview standard extended formal treatment meta-d′. However, whereas meta-d′ characterizes observer’s in tracking overall accuracy, it may sometimes be interest to assess metacognition particular kind behavioral response. For instance, perceptual task, wish characterize separately reports stimulus presence absence. discuss the computing such “response-specific” corresponding Matlab code. This approach potentially offers alternative explanation data that are typically taken support unequal variance (UV-SDT) model. We demonstrate simulated generated from UV-SDT can fit by equal model positing different ability each response, likewise latter captured UV-SDT. ambiguity entails caution is needed interpreting processes underlying relative operating characteristic (ROC) curve properties. Type 1 ROC curves combining type 2 judgments, traditionally interpreted terms low-level (UV), high-level instead (response-specific metacognition). Similarly, differences area under response-specific reflect influence (UV) rather than processes.