Neuroplastic effects of combined computerized physical and cognitive training in elderly individuals at risk for dementia: an eLORETA controlled study on resting states

Abstract

The present study investigates whether a combined cognitive and physical training may induce changes in the cortical activity as measured via electroencephalogram (EEG) and whether this change may index a deceleration of pathological processes of brain aging. Seventy seniors meeting the clinical criteria of mild cognitive impairment (MCI) were equally divided into 5 groups: 3 experimental groups engaged in eight-week cognitive and/or physical training and 2 control groups: active and passive. A 5-minute long resting state EEG was measured before and after the intervention. Cortical EEG sources were modelled by exact low resolution brain electromagnetic tomography (eLORETA). Cognitive function was assessed before and after intervention using a battery of neuropsychological tests including the minimental state examination (MMSE). A significant training effect was identified only after the combined training scheme: a decrease in the post- compared to pre-training activity of precuneus/posterior cingulate cortex in delta, theta, and beta bands. This effect was correlated to improvements in cognitive capacity as evaluated by MMSE scores. Our results indicate that combined physical and cognitive training shows indices of a positive neuroplastic effect in MCI patients and that EEG may serve as a potential index of gains versus cognitive declines and neurodegeneration. This trial is registered with ClinicalTrials.gov Identifier NCT02313935.

Figure 1

Flow of participants within the 3 experimental and 2 control groups.

Figure 2

Grand average of eLORETA solutions (i.e., CDR at PCu/PCC voxels at P < 0.05, corrected) modelling the EEG source for delta band in the LLM group on the corresponding axial (left view) and sagittal (right view) generic MRI slices. The left side of the maps (left view) corresponds to the left hemisphere. The power estimate was scaled based on the averaged maximum value indicated in the scale bar. Similar illustrations but of fewer voxels apply for the theta, beta 1, and beta 2 bands.

Figure 3

Visualization of the negative correlation of delta (r = −0.546, P = 0.043) (a) and theta (r = −0.633, P = 0.015) (b) bands among the MMSE score post- to pre-difference of each participant and the CDR post- to pre-difference of PCu/PCC activity that was statistically significant at P < 0.05, corrected.