In search of new protocols of neurofeedback: Independent components of event-related potentials

Abstract

In this article we present a method for decomposing individual multichannel event-related potentials (ERPs) into functionally meaningful components by means of spatial filtering. The spatial filters are based on topographies of components obtained by application of Inde-pendent Component Analysis (ICA) to a large collection (n= 297) of individual ERPs in the paired GO/NOGO task. sLORETA has been used for depicting neuronal generators of independent components (ICs). The ICs are divided into sensory-related (visual N1 and N170) and executive components presumably associated with engagement operation (P3b), action suppression (P3 NOGO) and conflict monitoring (P4 NOGO) operations. In a pilot study on 10 healthy subjects the feasibility of ICA/ERP-based neurofeedback approach has been tested. A neurofeedback protocol was implemented to test the amplitude training of the P3b independent component generated in the parietal cortex. The effect of 20 min of neurofeedback on ERPs in healthy subjects is presented. The amplitude of the P3b component did not change significantly during the training session; however, the amplitude of another component named Slow Positive Wave statistically decreased during both the training and sham conditions. We believe that this change may be sufficiently significant to warrant additional research, as it may hold promise for alternative treatments for some psychiatric illnesses. © Taylor & Francis Group, LLC.