This proposal aims to explore the decoding of human brain activities using EEG signals for Brain Computer Interfaces (BCI) by utilizing a multi-view spatiotemporal hierarchical deep learning. Decoding of motor movement and decoding of visual information from the human brain are the two most common decoding problems in BCI. In this proposal, the PI would like to explore the decoding of visual information from human brain utilizing EEG signals. EEG data is a record of electrical activity of the brain over a period of time. EEG signals have been used in diagnostics of neurological disorders as well as in brain computer interfaces. Analysis and the recognition of EEG signals have been studied for decades. The common approach in the analysis and the classification of EEG signals is based on one-dimensional (1D) time series input. In this project, the PI will explore the transformation of 1D temporal EEG signals into 2D spatiotemporal EEG image sequences for feature extraction and recognition. Spatiotemporal EEG image sequences will be explored based on the views from different planes. According to recent studies in computer vision, Deep Learning is considered the state-of-the-art method for image recognition. The PI's previous studies have shown that hierarchical deep learning improves the recognition accuracy further. Therefore, the PI will explore the use of multi-view hierarchical deep learning methods for the classification of 2D spatiotemporal image sequences. Recurrent Neural Networks have been known for their great performances in the prediction and classification of the time series data. Thus, in the proposed hierarchical learning scheme, the PI will explore the use of Convolutional Neural Network (CNN) with Long-Short Term Memory (LSTM) within the proposed hierarchical learning scheme.