Signal Detection Theory

A statistical framework used to measure the accuracy of a system or person in distinguishing between the presence and absence of a target signal amid noise. In accessibility and assistive technology research, Signal Detection Theory is used to evaluate how well detection systems (such as activity monitors, gesture recognisers, or automated accessibility checkers) perform by quantifying their sensitivity (ability to correctly detect when a target is present) and response bias (tendency to favour yes or no responses). Key metrics include hit rate, false alarm rate, sensitivity index (A' or d'), and response bias (B"D or beta). SDT is particularly useful in evaluating assistive technologies because it separates a system's discriminative ability from its decision threshold, providing a more nuanced picture of performance than simple accuracy percentages.