Title:Current Issues in the Use of fMRI-Based Neurofeedback to Relieve Psychiatric Symptoms
Volume: 21
Issue: 23
Author(s): Thomas Fovet, Renaud Jardri and David Linden
Affiliation:
Keywords:
Machine learning, neurofeedback, pattern recognition, psychiatric disorder, real-time fMRI, self-efficacy.
Abstract: fMRI-based neurofeedback (fMRI-NF) is a non-invasive technique that allows participants to achieve
control of their own brain activity using the real-time feedback of the activity (measured indirectly based on the
BOLD signal) of a particular brain region or network. The feasibility of fMRI-NF in healthy subjects has been
documented for a variety of brain areas and neural systems, and this technique has also been proposed for the
treatment of psychiatric disorders in recent years. Through a systematic review of the scientific literature this paper
probes the rationale and expected applications of fMRI-NF in psychiatry, discusses issues that must be addressed in
the use of this technique to treat mental disorders. Six relevant references and five ongoing studies were identified
according to our inclusion criteria. These studies show that in most psychiatric disorders (major depressive disorder,
schizophrenia, personality disorders, addiction), patients are able to learn voluntary control of the neuronal activity of the targeted
brain region(s). Interestingly, in some cases, this learning is associated with clinical improvement, showing that fMRI-NF can potentially
be developed into a therapeutic tool. However, only low-level evidence is available to support the use of this relatively new technique in
clinical practice. Notably, no randomized, controlled trial is currently available in this field of research. Finally, methodological issues
and clinical perspectives (especially the potential use of pattern recognition in fMRI-NF protocols) are discussed.
