Table 3.
EEG studies of cognition.
| References | N, stroke patients | Controls | Stroke acuity at EEG collection | EEG data collection | N, channels | State in EEG collected | EEG connectivity technique | Networks/Regions investigated | Collected clinical scores | Main findings |
|---|---|---|---|---|---|---|---|---|---|---|
| Gur et al. (24) | 200 | No | Acute | Cross-sectional | 18 | Resting | Qualitative EEG | Bilateral hemispheres | Presence of clinically diagnosed dementia | Increased slow wave activity is an indicator of subsequent cognitive decline. |
| Scleiger et al. (59) | 20 | No | Acute | Cross-sectional | 8 | Resting | PSA | Bifrontal | FIM with cognitive subset | Frontal DAR and global, relative alpha-band activity positively correlated with cognitive outcomes. |
| Song et al. (60) | 105 | No | Chronic | Cross-sectional | 16 | Resting | BRF and relative δ, θ, α, and β band power | Bilateral hemispheres | MOCA | The risk of developing cognitive impairment is 14 times higher for those with low BRF power. |
| Aminov et al. (61) | 24 | No | Acute | Cross-sectional | 1 | Resting | DAR and DTR | FP1 | MOCA | DTR within 24 h is the predictor of MOCA scores at 90 days. |
| Swatridge et al. (62) | 9 | No | Chronic | Longitudinal | 64 | Behavioral-triggered | Event related potential (P300) | Bilateral hemispheres | modified Eriksen Flanker (attention and inhibition) | P300 amplitude latency is shorter after exercise (increased cortical activity). No improvement in cognitive control. |
| Petrovic et al. (63) | 10 | Healthy | Chronic | Cross-sectional | 19 | Resting | PSA and αAVG | Four lateral frontal, and corresponding lateral posterior | MOCA | Frontal inter-hemispheric alpha activity may be a permanent consequence of asymptomatic cognitive impairment. |
| Romeo et al. (64) | 38 | Healthy | Subacute to chronic | Cross-sectional | 64 | Resting | Resting state network analysis; temporal ICA for mapping in networks previously identified in fMRI | Bilateral hemispheres | FIM, attentional matrices | Tested feasibility of 10 min and map cognitive functions in resting state networks. Visuo-spatial and motor impairments were primarily associated with the dorsal attention network. |
| *Dubovik et al. (51) | 20 | Healthy | Subacute | Cross-sectional | 128 | Resting | Adaptive spatial filter and imaginary component of coherence was calculated as an index of FC. | Bilateral hemispheres | VF, VWM, SWM, NHPT, STREAM, FMA. | FC in contralesional hemisphere negatively correlated with cognitive performance. Decreases in alpha-band coherence between a given node and the rest of the brain predicted deficits, independent of anatomical lesions. Increased alpha activity in the right IFG areas correlated negatively with VF. |
*
Combined motor and cognition study.
NHPT, Nine Hole Peg test; STREAM, Stroke; Rehabilitation Assessment of Movement; VFF, Verbal phonetic fluency; VWM, verbal working memory; SMW, spatial working memory tests; ICA, independent component analysis; fMRI, functional magnetic resonance imaging; FIM, Functional Independence Measure; BRF, Background rhythm frequency; PSA, Power Spectral Analysis; DAR, Delta alpha ratio; DTR, delta/theta ratio; MOCA, Montreal Cognitive Assessment; αAVG, the weighted average of alpha frequency; Acute, 0–2 weeks after stroke; Subacute, 2weeks-6 months after stroke; Chronic, 6 months and beyond.