Spectral perturbations reflect the depth of cognitive processing for brain-computer interface systems

Current trends and needs in Human-Computer Interaction (HCI) are rapidly increasing; however the possibility to estimate user's state remains unchanged. The need to predict user's state and intentions, and to adapt accordingly in real time to her/his decisions could bring benefits in this dual interaction. Following this purpose, we evaluate the state of a user by means of cognitive processing. In this way, we study the feasibility of detecting the level of cognitive processing, in order to be used for controlling or modifying the interface of a HCI application, such as, an information seeking application. We study this feasibility by analyzing the event-related spectral perturbations evoked by two levels of cognitive processing.