An electrophysiological model of working memory performance. (October 2017)

• Record Type: Journal Article
• Authors: Baghdadi, Golnaz; Towhidkhah, Farzad; Rostami, Reza
• Is Part Of: Cognitive systems research. Volume 45(2017)
• Page Start: 1
• Page End: 16
• Publication Date: 2017-10
• Subjects: Working memory -- Attention -- Electrophysiological model -- Neuronal activity; Cognition -- Periodicals; Cognitive engineering (System design) -- Periodicals; Artificial intelligence -- Periodicals; 153.05
• DOI: 10.1016/j.cogsys.2017.04.005
• Languages: English
• ISSNs: 1389-0417

Abstract

Abstract: Working memory (WM) enables us to keep a limited amount of information in active mode. It is believed that attention refreshes necessary information in WM and prevents their forgetting. Despite a plethora of models offered, it is not fully understood that what factors may be involved in forgetfulness and in the required time for refreshing the information. In this study, an electrophysiological model of WM is proposed that consists of several resistor-capacitor units. Inspired of the "resource capacity theory, " attention as a limited source of energy refreshes the voltage level of these units. According to the "time-based resource sharing theory, " only one of these units is allowed to use the limited source of attention at each moment. The source of attention is shared between active units. This model mimics the pattern of several well-known observations of WM such as the recall interval, the word length, and the serial position effect. Some suggestions have been provided about influencing factors in WM performance. Model parameters give the ability of investigating the possible effect of some other factors on WM performance and also a probable prediction about how much information can we chunk?