A dynamic stimulus-driven model of signal detection.

• DOI: 10.1037/a0025191
• Publication Date: 2011-09-06T16:00:58Z
• Authors: Brandon M. Turner; Trisha Van Zandt; Scott Brown
• Keywords: Stimulus (psychology); Detection theory; Observer (physics)

Signal detection theory forms the core of many current models cognition, including memory, choice, and categorization. However, classic signal model presumes a priori existence fixed stimulus representations--usually Gaussian distributions--even when observer has no experience with task. Furthermore, requires to place response criterion along axis strength, without theoretical elaboration, this is independent observer's experience. We present dynamic, adaptive that addresses these 2 long-standing issues. Our describes how representation can develop from rough subjective prior thereby explains changes in performance over time. The structure also provides basis for decision does not require placement strength. simulations examine its behavior several experiments provide data test model. fit recognition memory discuss role feedback plays establishing representations.