Decoding Olfactory Stimuli in EEG Data using Nonlinear Features: A Pilot Study.

•Methods based on nonlinear features are proposed for decoding perceived odors from recorded EEG signals.•The proposed methodology can be employed both for subject-specific and cross-subject odor decoding.•Odor pleasantness decoder can be reliably trained with data from different participants.•The results may prove helpful in developing markers of olfactory decline associated with neurodegenerative diseases such as Alzheimer’s.