Neurofeedback (NF) is a research and clinical technique, characterized by live demonstration of brain activation to the subject. The technique has become increasingly popular as a tool for the training of brain self-regulation, fueled by the superiority in spatial resolution and fidelity brought along with real-time analysis of fMRI (functional magnetic resonance imaging) data, compared to the more traditional EEG (electroencephalography) approach. NF learning is a complex phenomenon and a controversial discussion on its feasibility and mechanisms has arisen in the literature. Critical aspects of the design of fMRI-NF studies include the localization of neural targets, cognitive and operant aspects of the training procedure, personalization of training, and the definition of training success, both through neural effects and (for studies with therapeutic aims) through clinical effects. In this paper, we argue that a developmental perspective should inform neural target selection particularly for pediatric populations, and different success metrics may allow in-depth analysis of NF learning. The relevance of the functional neuroanatomy of NF learning for brain target selection is discussed. Furthermore, we address controversial topics such as the role of strategy instructions, sometimes given to subjects in order to facilitate learning, and the timing of feedback. Discussion of these topics opens sight on problems that require further conceptual and empirical work, in order to improve the impact that fMRI-NF could have on basic and applied research in future.

No abstract

Electrophysiological recordings of neuronal activity show spontaneous and task-dependent changes in their frequency-domain power spectra. These changes are conventionally interpreted as modulations in the amplitude of underlying oscillations. However, this overlooks the possibility of underlying transient spectral ‘bursts’ or events whose dynamics can map to changes in trial-average spectral power in numerous ways. Under this emerging perspective, a key challenge is to perform burst detection, i.e. to characterise single-trial transient spectral events, in a principled manner. Here, we describe how transient spectral events can be operationalised and estimated using Hidden Markov Models (HMMs). The HMM overcomes a number of the limitations of the standard amplitude-thresholding approach to burst detection; in that it is able to concurrently detect different types of bursts, each with distinct spectral content, without the need to predefine frequency bands of interest, and does so with less dependence on a priori threshold specification. We describe how the HMM can be used for burst detection and illustrate its benefits on simulated data. Finally, we apply this method to empirical data to detect multiple burst types in a task-MEG dataset, and illustrate how we can compute burst metrics, such as the task-evoked timecourse of burst duration.