Skip to main content

Table 4 EEG Features Predicting Trial Accuracy in SZ

From: Machine learning identification of EEG features predicting working memory performance in schizophrenia and healthy adults

Location

WM stage

Frequency

Feature Weight

Central

Encode

gamma

−1.169

Frontal

Encode

gamma

−0.916

Central

Retrieve

beta

0.704

Central

Retain

theta 1

−0.611

Central

Retrieve

theta 1

−0.601

Frontal

Retrieve

gamma

−0.600

Central

Encode

alpha

0.409

Occipital

Retrieve

theta 1

−0.371

Occipital

Retain

beta

−0.350

Central

Baseline

theta 1

−0.204

Frontal

Retain

theta 1

−0.168

Frontal

Encode

theta 1

−0.032

Frontal

Encode

theta 2

−0.029

Occipital

Retain

gamma

−0.001

  1. Features extracted by 1-norm SVM to classify correct vs. incorrect trials in SZ. Model based on simultaneous entry of 60 EEG features with correct trials labeled 1 and incorrect labeled −1. All other features weighted at 0