Abnormal EEG signal energy in the elderly: a wavelet analysis of event-related potentials during a Stroop task

Abstract

Background: Previous work showed that elderly with excess in theta activity in their resting state electroencephalogram (EEG) are at higher risk of cognitive decline than those with a normal EEG. By using event-related potentials (ERP) during a counting Stroop task, our prior work showed that elderly with theta excess have a large P300 component compared with normal EEG group. This increased activity could be related to a higher EEG signal energy used during this task. New method: By wavelet analysis applied to ERP obtained during a counting Stroop task we quantified the energy in the different frequency bands of a group of elderly with altered EEG. Results: In theta and alpha bands, the total energy was higher in elderly subjects with theta excess, specifically in the stimulus categorization window (258–516 ms). Both groups solved the task with similar efficiency. Comparison with existing methods: The traditional ERP analysis in elderly compares voltage among conditions and groups for a given time window, while the frequency composition is not usually examined. We complemented our previous ERP analysis using a wavelet methodology. Furthermore, we showed the advantages of wavelet analysis over Short Time Fourier Transform when exploring EEG signal during this task. Conclusions: The higher EEG signal energy in ERP might reflect undergoing neurobiological mechanisms that allow the elderly with theta excess to cope with the cognitive task with similar behavioral results as the normal EEG group. This increased energy could promote a metabolic and cellular dysregulation causing a greater decline in cognitive function.