Model study of target discrimination in concurrent auditory events. Issue 3 (22nd February 2022)

• Record Type: Journal Article
• Authors: Yihang, Du; Chao, Sun; Ke, Niu
• Is Part Of: Cognitive computation and systems. Volume 4:Issue 3(2022)
• Page Start: 294
• Page End: 303
• Publication Date: 2022-02-22
• Subjects: auditory selective attention (ASA) -- speech shaped noise -- target discrimination; Cognitive science -- Periodicals; Artificial intelligence -- Periodicals; Neurosciences -- Periodicals; Computer science -- Periodicals; Neurosciences; Computer science; Cognitive science; Artificial intelligence; Periodicals; Electronic journals; 006.3
• DOI: 10.1049/ccs2.12052
• Languages: English
• ISSNs: 2517-7567

Abstract

Abstract: With the development of automation technology in human‐machine interface, the traditional visual display no longer met the needs of complicated control systems, while auditory display has become an important part of human‐machine interface design. Analysing the auditory attention selection behaviours in concurrent auditory events is essential for the design and evaluation of the auditory interface. Listeners tend to pay attention to auditory target with physical attributes which were more salient and informative with deep cognitive level. Listeners tend to focus on the important information in the concurrent auditory events. By combining the two tendencies of the selective attention mechanism, the binary Logit model was built to analyse human auditory selective attention (ASA) behaviour, and the auditory target's selection can be transformed into a probability event. The continuous speech‐shaped noise (SSN) was described by the concurrent auditory events scene, which is a typical target discrimination task for ASA. An experiment using the target discrimination task in SSN was conducted to verify the model. The results showed that the listener's attention behaviour was in good agreement with the proposed Logit model. The proposed ASA model is effective and reasonable, which is promising for use in behaviour analysis, auditory display optimization, and interface design.