Tunnustest

Tunnustest, literally "recognition test" in Finnish, is a term used in Finnish-language literature to describe assessment methods that measure recognition abilities. In psychology, tunnustest refers to recognition memory tasks in which participants decide whether items were previously encountered. Typical formats present a mix of old (studied) items and new items and record responses as hits, misses, false alarms, and correct rejections. From these data, researchers often compute accuracy and bias measures, and may apply signal detection theory metrics such as d-prime and criterion.

In computing and information retrieval, tunnustest can denote evaluation procedures for systems that must recognize or retrieve previously seen data, for example in image or text recognition. Common formats include binary yes/no judgments, multiple-choice selections, or forced-choice assignments, with performance summarized by accuracy, precision, recall, F1, and, in some cases, ROC-based measures.

Applications of tunnustest span cognitive psychology experiments, aging and clinical studies of memory, and benchmarking of recognition-based algorithms in AI. When used in research, careful design considerations include controlling stimulus sets, counterbalancing, and accounting for practice effects and ceiling effects.

Limitations include variability in definitions and translations of the term across studies, potential linguistic or cultural biases in stimuli, and the lack of a single standardized Finnish test under this name. See also recognition memory, memory tests, signal detection theory, and model evaluation for recognition tasks.