Adapted wavelet transform improves time-frequency representations: a study of auditory elicited P300-like event-related potentials in rats

P300 in schizophrenia: Then and now

The 1965 discovery of the P300 component of the electroencephalography (EEG)-based event-related potential (ERP), along with the subsequent identification of its alteration in people with schizophrenia, initiated over 50 years of P300 research in schizophrenia. Here, we review what we now know about P300 in schizophrenia after nearly six decades of research. We describe recent efforts to expand our understanding of P300 beyond its sensitivity to schizophrenia itself to its potential role as a biomarker of risk for psychosis or a heritable endophenotype that bridges genetic risk and psychosis phenomenology. We also highlight efforts to move beyond a syndrome-based approach to understand P300 within the context of the clinical, cognitive, and presumed pathophysiological heterogeneity among people diagnosed with schizophrenia. Finally, we describe several recent approaches that extend beyond measuring the traditional P300 ERP component in people with schizophrenia, including time-frequency analyses and pharmacological challenge studies, that may help to clarify specific cognitive mechanisms that are disrupted in schizophrenia. Moreover, we discuss several promising areas for future research, including studies of animal models that can be used for treatment development.

Vowel speech recognition from rat electroencephalography using long short-term memory neural network

Over the years, considerable research has been conducted to investigate the mechanisms of speech perception and recognition. Electroencephalography (EEG) is a powerful tool for identifying brain activity; therefore, it has been widely used to determine the neural basis of speech recognition. In particular, for the classification of speech recognition, deep learning-based approaches are in the spotlight because they can automatically learn and extract representative features through end-to-end learning. This study aimed to identify particular components that are potentially related to phoneme representation in the rat brain and to discriminate brain activity for each vowel stimulus on a single-trial basis using a bidirectional long short-term memory (BiLSTM) network and classical machine learning methods. Nineteen male Sprague-Dawley rats subjected to microelectrode implantation surgery to record EEG signals from the bilateral anterior auditory fields were used. Five different vowel speech stimuli were chosen, /a/, /e/, /i/, /o/, and /u/, which have highly different formant frequencies. EEG recorded under randomly given vowel stimuli was minimally preprocessed and normalized by a z-score transformation to be used as input for the classification of speech recognition. The BiLSTM network showed the best performance among the classifiers by achieving an overall accuracy, f1-score, and Cohen’s κ values of 75.18%, 0.75, and 0.68, respectively, using a 10-fold cross-validation approach. These results indicate that LSTM layers can effectively model sequential data, such as EEG; hence, informative features can be derived through BiLSTM trained with end-to-end learning without any additional hand-crafted feature extraction methods.