A Multi-stage Theory of Neurofeedback Learning

Abstract

Neurofeedback is a training paradigm through which trainees learn to voluntarily influence their brain dynamics. Recent years have seen an exponential increase in research interest into this ability. How neurofeedback learning works is still unclear, but progress is being made by applying models from computational neuroscience to the neurofeedback paradigm. In this chapter, I will present a multi-stage theory of neurofeedback learning, which involves three stages, involving different neural networks. In stage 1, the system discovers the appropriate goal representation for increasing the frequency of positive feedback. This stage operates at a within-session timescale and is driven by reward-based learning, which updates fronto-striatal connections. Stage 2 operates on a timescale that covers multiple training sessions and is sensitive to consolidation processes. This stage involves updating striatal-thalamic and thalamo-cortical connections. Finally, after stages 1 and 2 have started, stage 3 may be triggered by the awareness of the statistical covariation between interoceptive and external feedback signals. When this awareness emerges, neurofeedback learning may speed up and its effect be maintained well after the conclusion of the training period. Research guided by this framework is described that consist of quantitative, qualitative, and computational methodologies. At present, the findings suggest that the framework is able to provide novel insights into the nature of neurofeedback learning and provides a roadmap for developing instructions that are designed to facilitate the likelihood of learning success.