EEG-Based Methods to Characterize Memorised Visual Space

Abstract

One second of memory maintenance was evaluated to determine EEG metrics ability to track memory load and its variations connected with the lateral presentation of objects in the visual hemi-field. An initial approach focused on features gathered from the N2pc time series to detect the memory load using ensemble learners. Conversely, the secondary approach employed a regularised support vector classifier to predict the area of N2pc event-related components, identifying 6 levels of memory load and stimulus location.