Intrusive mental states: from real-time decoding of neural activations associated with hallucinations to innovative treatments – INTRUDE

Neuropsychiatric disorders are the leading causes of disability and health costs in developed countries. Altogether, the direct costs (spending for treatment, hospitalizations and rehabilitation nationwide) and indirect costs of mental disorders (lost productivity at the workplace, school, and home due to premature death or disability) amount to 199 M$ of health spending worldwide (WHO, 2008). Schizophrenia (SCZ) is one of the most devastating neuropsychiatric disorders, with a prevalence of about 1% over lifetime. SCZ is characterized by the presence of hallucinations, delusions, and cognitive deficits. More than 25% of SCZ patients show no or incomplete responses to current pharmacological (antipsychotics) or psychological/remediation therapies for hallucinations. These refractory hallucinations have a deep impact on autonomy, life achievements and medical use, with complex multi-level social consequences.

After five decades without any significant therapeutic breakthrough, pharmaceutical industries recently cut down their research investments on neuropsychiatric disorders while these medical needs remain unmet. On the other hand, non-invasive neuromodulation recently proved to be a safe and effective treatment for drug-resistant hallucinations, but its benefits can only be maintained at the cost of time-consuming daily administrations, which have prevented its widespread use so far. In this context, the INTRUDE project takes a translational stance to create and validate an innovative non-pharmacological therapy for intrusive pathological mental states in psychiatry.

Real-time functional magnetic resonance imaging (rt-fMRI) is a novel brain imaging approach allowing for online feedbacks derived from mean brain activity in a priori regions. This technology, known as neurofeedback, already demonstrated its ability to alleviate chronic pain by training patients to voluntarily suppress brain activity in key regions. Hence, rt-fMRI and neurofeedback represent a significant step toward effective fMRI-based neurofeedback strategies to relieve invalidating subjective symptoms, such as psychotic symptoms by enhancing patient’s ability to control over intrusive thoughts. However, intrusive thoughts emerge from complex and transient patterns of activity distributed brain-wide, which are not captured by conventional neurofeedback approaches.

The INTRUDE project plans to address this issue by adapting multivariate decoding techniques to rt-fMRI. Offline multivariate decoders can already capture fine-grained spatial patterns of fMRI activity to accurately predict mental states, such as emerging perception or free choices. Using this innovative approach, the INTRUDE project aims at developing and validating a robust real-time multivariate decoding algorithm for hallucinations during fMRI acquisitions, assess its therapeutic benefits and tolerance in Schizophrenia patients, and probe its possible extension to other intrusive mental categories, such as obsessive thoughts in Obsessive Compulsive Disorders.