Hallucination brain networks can be self-regulated via neurofeedback

Hallucinations can occur in the healthy population, are clinically relevant and frequent symptoms in many neuropsychiatric conditions, and have been shown to mark disease progression in patients with neurodegenerative disorders where antipsychotic treatment remains challenging (Bernasconi et al., 2022). These challenges are especially prevalent in patients with Parkinson’s disease (PD) and Dementia with Lewy bodies (DLB). In a collaboration between the Blanke and the Van De Ville Lab’s at EPFL, Dhanis et al. developed a proof-of-concept method of a non-invasive therapy for hallucinations in PD: a new neurotechnology-based antipsychotic treatment. EPFL has previously developed an MRI-compatible robotic set-up to induce a specific clinically-relevant hallucination in the MRI scanner (Bernasconi & Blondiaux et al., 2021) and had identified brain network dynamics associated with this type of hallucination (Dhanis et al, 2022). In the current work, Dhanis et al., trained participants, over three days, to up-regulate occurrences of this hallucination network, leading to heightened sensitivity to robot-induced specific hallucinations, after the training. Moreover, participants who became sensitive and succeeded in fMRI-NF training, showed sustained and specific neural changes after training, characterized by increased hallucination network occurrences during induction and decreased hallucination network occurrences during a matched control condition. These data demonstrate that fMRI-NF modulates specific hallucination network dynamics and highlights the potential of fMRI-NF as a novel antipsychotic treatment in neurodegenerative disorders (PD, DLB) and schizophrenia. In on-going work, we have started applying this new therapy to patients with PD.