A promising neuroimaging method for evaluating brain activity and cognitive processes is electroencephalography (EEG). An essential step towards making the use of EEG more widely applicable and less dependent on experienced specialists is the automatic classification of these signals. This paper presents a preliminary study on the classification of brain signals associated with inhibition responses during a Stroop task using a one-dimensional convolutional neural network (1D-CNN). The EEG signals captured at \({\text{F}}_{p1}\) and \({\text{F}}_{p2}\) electrodes were used to classify between rest and Stroop task activity. We adopted an end-to-end learning approach, where raw data was directly fed into the model. To enhance the dataset, we utilized a sliding window technique as data augmentation. The signals have then been trained using three different CNN-based classifiers: 2-layer compact CNN (CCNN), 1-layer compact CNN (CCNN), and standard CNN. Finally, the performance of these models was evaluated using accuracy as the metric. The two-layer CCNN architecture was found to have the highest accuracy, achieving nearly 90% accuracy for both \({\text{F}}_{p1}\) and \({\text{F}}_{p2}\) electrodes in classifying rest and inhibition response task signals. The results demonstrated that the classifier performed well in classifying inhibition response signals.