Error-Related Potential Classification Through the Use of the Detectivity Parameter

In recent years, in the context of “Human-Robot Interaction” (HRI), research on Error-related Potentials (ErrPs), a specific Event-Related Potential (ERP) used in electroen-cephalogram (EEG) analysis, has gained a growing attention in developing robotic prostheses that may learn users' needs and adjust to time-specific demands. This work aims to develop an innovative method to classify ErrP offline. This new method uses a parameter called a detectogram (used in predictive diagnostics) to identify potential (ErrPs). The analysis is carried out using different EEG datasets, which has been pre-processed. The detectogram is calculated, analyzed, and compared with classical classification method already known in the literature.