<p>The cognitive workload (CWL) triggers neural activity, which is crucial for understanding the brain’s response to mental stress or stimuli that induce stress. Electroencephalogram (EEG) signals were collected from a mental arithmetic task (MAT), simultaneous task EEG workload datasets, and segmented into 4-s intervals. These segmented signals were then transformed into images using time–frequency conversion methods (TF) called superlet transform (SLT). The resulting TF images were fed into convolutional neural networks (CNNs), such as VGG16, ResNet50, Xception, EfficientNetB0, AlexNet, GoogLeNet, SqueezeNet, VGG16 + LSTM, VGG16 + BiLSTM, FNet, gMLP, and multilayer perceptrons (MLP) mixer. CNN models were trained using the Adam optimizer to detect cognitive load. The preprocessing involved normalization, and scaling in both phases. Among the models tested, the SLT-based TFEEG with the MLP Mixer outperformed other CNN architectures. It helps reduce overfitting and vanishing gradients, enhances performance with new data, improves GPU acceleration, and reduces computational cost due to its simpler architecture. However, the SLT effectively handles non-stationary data through its adaptive multiresolution approach, making it ideal for EEG analysis. The proposed SLT + MLP Mixer achieved an accuracy of 98.69%, precision of 98.66%, sensitivity of 97.34%, Cohen’s kappa of 92.25%, Jaccard index of 98.37%, MCC of 95.99%, specificity of 97.35%, and F-score of 97.88% for MAT datasets, respectively. These findings can enhance real-time cognitive load monitoring, benefiting areas such as personalized learning, mental health, and stress management by detecting CWL to improve performance and reduce stress in critical situations.</p>