Confidence Representation of Perceptual Decision by EEG and Eye Data in a Random Dot Motion Task. (15th May 2019)

• Record Type: Journal Article
• Authors: Vafaei Shooshtari, Shirin; Esmaily Sadrabadi, Jamal; Azizi, Zahra; Ebrahimpour, Reza
• Is Part Of: Neuroscience. Volume 406(2019)
• Page Start: 510
• Page End: 527
• Publication Date: 2019-05-15
• Subjects: EEG electroencephalography -- RDM random dot motion -- DDM drift diffusion model -- ERP event related potential ERP -- Pe error positivity -- RT reaction time -- ICA independent component analysis -- LIP lateral intraparietal cortex -- FEF frontal eye field -- CSD current source density -- SEM standard error of the mean -- SVM support vector machine; decision making -- confidence -- EEG -- eye data -- time-dependent analysis; Neurochemistry -- Periodicals; Neurophysiology -- Periodicals; Neurology -- Periodicals; Neurochimie -- Périodiques; Neurophysiologie -- Périodiques; Neurochemistry; Neurophysiology; Electronic journals; Periodicals; Electronic journals; 612.8
• DOI: 10.1016/j.neuroscience.2019.03.031
• Languages: English
• ISSNs: 0306-4522

Abstract

Abstract: The Confidence of a decision could be considered as the internal estimate of decision accuracy. This variable has been studied extensively by different types of recording data such as behavioral, electroencephalography (EEG), eye and electrophysiology data. Although the value of the reported confidence is considered as one of the most important parameters in decision making, the confidence reporting phase might be considered as a restrictive element in investigating the decision process. Thus, decision confidence should be extracted by means of other provided types of information. Here, we proposed eight confidence related properties in EEG and eye data which are significantly descriptive of the defined confidence levels in a random dot motion (RDM) task. As a matter of fact, our proposed EEG and eye data properties are capable of recognizing more than nine distinct levels of confidence. Among our proposed features, the latency of the pupil maximum diameter through the stimulus presentation was established to be the most associated one to the confidence levels. Through the time-dependent analysis of these features, we recognized the time interval of 500–600 ms after the stimulus onset as an important time in correlating features to the confidence levels. Highlights: The DDM analysis showed that the confidence levels were correlated to the model parameters, especially the drift rate. Our exploration introduced four EEG and four eye data properties which were able to predict decision confidence. Analysis of the proposed properties highlighted the importance of 500–600 ms after the onset of the stimulus in the confidence formation.