Principal component analysis of auditory ERPs discriminates mild cognitive impairment and Alzheimer’s disease from normal aging

Background Sensory gating refers to an inhibitory mechanism of the brain that protects higher cortical centers from being flooded with repetitive, redundant sensory inputs. Sensory gating is modulated by cholinergic transmission, the deterioration of which is a prominent neurochemical feature of Alzheimer’s disease (AD), the most common form of dementia in the elderly. The study investigated whether alterations in inhibitory gating, assessed by event‐related potentials (ERPs) in response to repeated auditory stimuli, can discriminate AD and mild cognitive impairment (MCI) from normal aging. Method Event‐related electroencephalogram was recorded from 38 AD patients, 37 MCI patients, and 38 elderly cognitively normal control (NC) participants in a conditioning–testing paradigm. ERP waveforms, stimulus‐locked changes in power, and inter‐trial phase clustering (ITPC) were extracted and submitted to a covariance‐based principal component analysis (PCA) to yield data‐driven component measures. Result Non‐parametric cluster‐based permutation tests revealed statistically significant differences for ERP waveforms in separating AD, for power in separating AD and MCI, and for ITPC in separating AD, MCI, and NC from the other two groups pooled together. Conclusion The study confirms that alterations in auditory sensory gating can discriminate between AD, MCI, and normal aging. Auditory ERPs might serve as an objective, non‐invasive, and cost‐effective method for evaluating cognitive decline suggestive of AD and to monitor disease progression, therapeutic efficacy, or effects of novel drug candidates in clinical trials.