Cortical connectivity and memory performance in cognitive decline: A study via graph theory from EEG data

F. Vecchio, F. Miraglia, D. Quaranta, G. Granata, R. Romanello, C. Marra, P. Bramanti, P. M. Rossini

Research output: Contribution to journalArticlepeer-review

Abstract

Functional brain abnormalities including memory loss are found to be associated with pathological changes in connectivity and network neural structures. Alzheimer's disease (AD) interferes with memory formation from the molecular level, to synaptic functions and neural networks organization. Here, we determined whether brain connectivity of resting-state networks correlate with memory in patients affected by AD and in subjects with mild cognitive impairment (MCI). One hundred and forty-four subjects were recruited: 70 AD (MMSE Mini Mental State Evaluation 21.4), 50 MCI (MMSE 25.2) and 24 healthy subjects (MMSE 29.8). Undirected and weighted cortical brain network was built to evaluate graph core measures to obtain Small World parameters. eLORETA lagged linear connectivity as extracted by electroencephalogram (EEG) signals was used to weight the network. A high statistical correlation between Small World and memory performance was found. Namely, higher Small World characteristic in EEG gamma frequency band during the resting state, better performance in short-term memory as evaluated by the digit span tests. Such Small World pattern might represent a biomarker of working memory impairment in older people both in physiological and pathological conditions.

Original languageEnglish
Pages (from-to)143-150
Number of pages8
JournalNeuroscience
Volume316
DOIs
Publication statusPublished - Mar 1 2016

Keywords

  • Alzheimer and MCI
  • EEG and alpha band
  • eLORETA
  • Functional connectivity
  • Graph theory
  • Memory

ASJC Scopus subject areas

  • Neuroscience(all)

Fingerprint Dive into the research topics of 'Cortical connectivity and memory performance in cognitive decline: A study via graph theory from EEG data'. Together they form a unique fingerprint.

Cite this