Deep Learning (DL) has recently shown promising classification performance in Electroencephalography (EEG) in many different scenarios. However, the complex reasoning of such models often prevent the user to explain their classification abilities. Attention, one of the most
recent and influential ideas in DL, allows the models to learn which portions of the data are relevant to the final classification output. In this
work, we compared three attention-enhanced DL models, the brand-new InstaGATs , an LSTM with attention and a CNN with attention. We used these models to classify normal and abnormal, including artifactual and pathological, EEG patterns in three different datasets. We achieved the state of the art in all classification problems, regardless the large variability of the datasets and the simple architecture of the attention-enhanced models. Additionally, we proved that, depending on how the attention mechanism is applied and where the attention layer is located in the model, we can alternatively leverage the information contained in the time, frequency or space domain of the EEG dataset. Therefore, attention represents a promising strategy to evaluate the quality of the EEG information, and its relevance for classification, in different real-world scenarios.