Memory-related EEG markers in aMCI: beta time-frequency activity during an omitted tone task

Objective Mild Cognitive Impairment (MCI) subclinical pictures are deemed as prodromal stages of Alzheimer’s disease (AD) or other dementia. In particular, amnestic MCI (aMCI) shows the strongest association with AD and is primarily characterized by memory impairments, whose severity associate with the likelihood and steepness of progression to dementia. According to the attention deficit hypothesis, aMCI symptoms may ground on a global attention network and working memory decline. Monitoring those domains may then usefully help in tracking and predicting clinical development. The present study aims at investigating the potential of attention-related beta EEG activity as markers of aMCI. Participants To investigate task-related beta responses we designed an omitted tone task, where participants had to explicitly orient their attention to estimate repeated intervals in a series of tones and signal when a tone is omitted. 15 aMCI patients and 15 healthy controls took part in the study. Event-related spectral perturbation (ERSP) and inter-trial coherence (ITC) were computed to assess time-frequency beta EEG patterns collected during the task. Results Bootstrap statistics on processed electrophysiological data highlighted that both groups presented event-related desynchronization during the tone condition and event-related synchronization during the omitted tone condition, but their responses differed in terms of spatial-temporal pattern. Further, analyses highlighted peculiar coherence indices during both conditions in aMCI patients. Conclusion Given their functional correlates, ERSPs and ITC modulations may mark the alteration of cortical information processing and exchange in attention/working-memory networks accompanied by a pathological enhancement of sensory cortical excitability even in the prodromal stage of AD. Beta EEG responses measured during relevant cognitive tasks might then act as valuable clinical markers and might inform on the state of pathology-specific cognitive difficulties and of supporting cortical networks.