I am working with classification of people with ADHD using EEG signals, so I will answer you based on my work.
1.- Look for a dataset, where the desired labels come, (right-left move), exist some plase like:
mendeley dataset, UCI Machine Learning Repository, kaggle datset and so on.
2.-Look for previous work and the techniques they used to solve the problem, google Schoolar, Springer and so on.
2.-It is necessary to pre-process the data (filters, ICA), extract the most relevant characteristics (Continous Wavelet Transform among others).
3.- There are techniques that allow you to correctly classify SVM, LDA, ResNet and other techniques
