Signal Detection Analysis

To calculate d-prime and bias for yes/no recognition memory data, you need hit and false alarm rates from confusion matrices. This Skill computes sensitivity metrics like d' alongside bias measures such as c, c', and beta for accurate cognitive data interpretation.

Signal detection theory applies to 2AFC and rating scale paradigms by estimating ROC-based metrics like da and Az. This Skill provides unified guidance for computing these sensitivity measures and optional unequal-variance modeling across perceptual and memory tasks.

To handle extreme proportions in signal detection metrics, you must apply specialized corrections to avoid infinite z-scores. This Skill provides built-in routines for extreme-proportion corrections alongside zROC-based estimation for reliable sensitivity analysis.

Signal detection analysis can be used for eyewitness identification and clinical decision making by separating sensitivity from response criterion. This Skill supports these paradigms, allowing researchers to quantify discriminability accurately across diagnostic and cognitive domains.

Computing ROC and d-prime metrics requires data structured as confusion matrices or rating distributions from observed responses. This Skill processes these inputs to generate sensitivity estimates and bias metrics for perceptual detection tasks.