A customizable, multi-parametric auditory display technique for theta-alpha neurofeedback training

Advances in physiological measurement systems allowed equipment that used to be exclusive to medical or research centers to be acquired and used by virtually anyone. The Electroencephalogram is a good example of how these new technologies are getting closer to the non-scientific community, with portable and affordable devices like the Enobio and the Emotiv EPOC brain-computer interfaces. Together with advances in fields like behavioral and cognitive neurosciences, and psychology, we now can easily access complex data that gives us unparalleled insight of our brain and body states. On the other hand, the field of Auditory Display is a relatively young one but has nonetheless been successfully applied to a great group of very diverse research fields (data exploration, musical composition, interaction design, and for studying medical conditions to cite a few). Being perfectly suitable for displaying complex physiological data, it has been gaining much attention in last decades, especially when applied to monitoring and biofeedback systems. One important issue that arises from this situation is the fact that in such an interdisciplinary field, sonification experts have to design mappings for many different situations and users.

In this master thesis we aim at investigating how the use of multi-parameterization combined with the user customization of such mappings can play a role in the field of auditory neurofeedback. In order to explore this problem we developed a system capable of generating a broad range of sonifications from EEG signals and used it to explore these concepts in the well-established Theta-Alpha Neurofeedback Training paradigm. We analyze the subjective and physiological response from the subjects in an experimental scenario involving customizable natural soundscapes and musical elements as an auditory display of brain activity in specific frequency bands.

Results show that multi-parameterization seem to have a positive influence in the neurofeedback scenario and can be a promising direction to pursue while usercustomization is desired but still require some improvements to become feasible.