Signaalinhavaitsemisteoria

Signaalinhavaitsemisteoria, known in English as Signal Detection Theory (SDT), is a framework used to analyze the ability of observers to distinguish a signal from background noise. It is widely applied in fields such as psychology, neuroscience, engineering, and medicine. The core idea of SDT is that a decision to detect a signal is not solely based on the strength of the signal itself but also on the observer's internal response criterion.

The theory posits that every observation involves two components: the signal and noise. The observer's task is to decide whether a signal is present or absent. SDT quantures this process by considering four possible outcomes: a hit (correctly identifying a signal), a miss (failing to identify a signal when it is present), a false alarm (identifying a signal when it is absent), and a correct rejection (correctly identifying the absence of a signal).

Two key metrics are derived from SDT: sensitivity and criterion. Sensitivity, often denoted as d' (d-prime), measures how well the observer can discriminate between the signal and the noise, independent of their response bias. A higher d' indicates better discriminability. The criterion, often denoted as beta (β) or c, reflects the observer's bias or willingness to report a signal. A liberal criterion leads to more hits but also more false alarms, while a conservative criterion leads to fewer false alarms but also fewer hits. By analyzing the proportions of hits and false alarms, researchers can quantify both an observer's perceptual ability and their decision-making strategy.