Quantification of event-related desynchronization/synchronization at low frequencies in a semantic memory task

Although several techniques have been developed for the visualization of EEG event-related desynchronization/synchronization (ERD/ERS) in both time and frequency domains, none of the quantification methods takes advantage of the time and frequency resolution at the same time. Existing techniques for the quantification of the ERD/ERS changes compute the average EEG power increase/decrease relative to certain reference value, over fixed time intervals and/or frequency bands (either fixed or individualized). Inaccuracy in the computation of these frequency bands (where the process is actually measured) in combination with the averaging process over time may lead to errors in the computation of any ERD/ERS quantification parameter. In this paper, we present a novel method for the automatic, individual and exact quantification of the most significant ERD/ERS region within a given window of the time-frequency domain. The method is exemplified by quantifying the ERS at low frequencies of 10 subjects performing a semantic memory task, and compared with existing techniques.