Advances in signal processing and machine learning have expedited electroencephalogram (EEG)-based emotion recognition research, and numerous EEG signal features have been investigated to detect or characterize human emotions. However, most studies in this area have used relatively small monocentric data and focused on a limited range of EEG features, making it difficult to compare the utility of different sets of EEG features for emotion recognition. This study addressed that by comparing the classification accuracy (performance) of a comprehensive range of EEG feature sets for identifying emotional states, in terms of valence and arousal. The classification accuracy of five EEG feature sets were investigated, including statistical features, fractal dimension (FD), Hjorth parameters, higher order spectra (HOS), and those derived using wavelet analysis. Performance was evaluated using two classifier methods, support vector machine (SVM) and classification and regression tree (CART), across five independent and publicly available datasets linking EEG to emotional states: MAHNOB-HCI, DEAP, SEED, AMIGOS, and DREAMER. The FD-CART feature-classification method attained the best mean classification accuracy for valence (85.06%) and arousal (84.55%) across the five datasets. The stability of these findings across the five different datasets also indicate that FD features derived from EEG data are reliable for emotion recognition. The results may lead to the possible development of an online feature extraction framework, thereby enabling the development of an EEG-based emotion recognition system in real time.

Heightened pain sensitivity, the amount of pain experienced in response to a noxious event, is a known risk factor for development of chronic pain. We have previously reported that pain-free, sensorimotor peak alpha frequency (PAF) is a reliable biomarker of pain sensitivity for thermal, prolonged pains lasting tens of minutes. To test whether PAF can provide information about pain sensitivity occurring over clinically relevant timescales (i.e., weeks), EEG was recorded before and while participants experienced a long-lasting pain model, repeated intramuscular injection of nerve growth factor (NGF), that produces progressively developing muscle pain for up to 21 days. We demonstrate that pain-free, sensorimotor PAF is negatively correlated with NGF pain sensitivity; increasingly slower PAF is associated with increasingly greater pain sensitivity. Furthermore, PAF remained stable following NGF injection, indicating that the presence of NGF pain for multiple weeks is not sufficient to induce the PAF slowing reported in chronic pain. In total, our results demonstrate that slower pain-free, sensorimotor PAF is associated with heightened sensitivity to a long-lasting musculoskeletal pain and also suggest that the apparent slowing of PAF in chronic pain may reflect predisease pain sensitivity. NEW & NOTEWORTHY Pain sensitivity, the intensity of pain experienced after injury, has been identified as an important risk factor in the development of chronic pain. Biomarkers of pain sensitivity have the potential to ease chronic pain burdens by preventing disease emergence. In the current study, we demonstrate that the speed of pain-free, sensorimotor peak alpha frequency recorded during resting-state EEG predicts pain sensitivity to a clinically-relevant, human model of prolonged pain that persists for weeks.