Signal Detection Theory Explorer

An equal-variance SDT model. The blue curve is noise; the orange curve is signal-plus-noise, separated by sensitivity d′. The observer says “signal” for anything right of the criterion — drag it and watch hits and false alarms trade off while d′ stays fixed. Bias c measures how far the criterion sits from the neutral point between the two distributions. In Model mode, set d′ and c directly; in Data mode, enter observed hit and false-alarm rates and the tool recovers the model that implies them.