Strategies to solve the reverse inference fallacy in future MRI studies of schizophrenia: a review

Abstract

Few advances in schizophrenia research have been translated into clinical practice, despite 60 years of serum biomarkers studies and 50 years of genetic studies. During the last 30 years, neuroimaging studies on schizophrenia have gradually increased, partly due to the beautiful prospect that the pathophysiology of schizophrenia could be explained entirely by the Human Connectome Project (HCP). However, the fallacy of reverse inference has been a critical problem of the HCP. For this reason, there is a dire need for new strategies or research "bridges" to further schizophrenia at the biological level. To understand the importance of research "bridges," it is vital to examine the strengths and weaknesses of the recent literature. Hence, in this review, our team has summarized the recent literature (1995-2018) about magnetic resonance imaging (MRI) of schizophrenia in terms of regional and global structural and functional alterations. We have also provided a new proposal that may supplement the HCP for studying schizophrenia. As postulated, despite the vast number of MRI studies in schizophrenia, the lack of homogeneity between the studies, along with the relatedness of schizophrenia with other neurological disorders, has hindered the study of schizophrenia. In addition, the reverse inference cannot be used to diagnose schizophrenia, further limiting the clinical impact of findings from medical imaging studies. We believe that multidisciplinary technologies may be used to develop research "bridges" to further investigate schizophrenia at the single neuron or neuron cluster levels. We have postulated about future strategies for overcoming the current limitations and establishing the research "bridges," with an emphasis on multimodality imaging, molecular imaging, neuron cluster signals, single transmitter biomarkers, and nanotechnology. These research "bridges" may help solve the reverse inference fallacy and improve our understanding of schizophrenia for future studies.

A Narrative Review of Clinical researches of Acupuncture treatment for Depression using Neuroimaging method: Focusing on SCI papers

“…In the clinical practices of traditional Korean medicine, most depressive patients come to the hospital while taking antidepressants. In our opinion, there would be some demand for traditional Korean medicine to make up for the lack of therapeutic effect in conventional Alzheimer's disease 16) , depression 17) , schizophrenia 18) , Parkinson's disease 19) .…”

Two Cases of Schizophrenia Treated with Individualized Homoeopathy

“…Schizophrenia is a chronic psychiatric disorder with a heterogeneous genetic and neurobiological background that is expressed as a combination of psychotic symptoms, and motivational and cognitive dysfunctions [1]. Schizophrenia initially termed as 'Dementia praecox' by Kraepelin et al in 1877 A.D is characterized by hallucinations, delusions, and impaired cognitive abilities [2]. Later Eugen Bleuler coined the term 'Schizophrenia' meaning 'Splitting of mind [3].…”

“…Lateral ventricle enlargement is the most common finding in patients with schizophrenia, followed by third ventricle enlargement [2]. In humans, greater expression of C4A in the brain is associated with an increased risk of schizophrenia [11].…”

In the era of whole-brain mapping for the exploration of mental disorders, we need to rethink our methods of rodent model establishment

“…Other hypotheses, such as those involving inflammation, neuroimmunity, vitamin and trace element deficiencies, and intestinal flora, as well as those based on multi-omics research, are being integrated into research in this field [ 4 ]. As pointed out by Zhuo et al, however, the fact that “the reverse inference is not tenable” remains a key issue in psychiatric research [ 5 ].…”

“…However, the computational complexity of human brains is billions of times that of rodent brains, and computing speeds will not be able to accommodate such complexity in the next decade. Despite much progress, such techniques are not currently suitable for the exploration of new methods to improve patient prognoses [ 5 ].…”