Psychiatric diagnosis is descriptive; some of the difficulties generated by it were introduced in ‘The Research Agenda for DSM-V’
For example “….it can be concluded that the field of psychiatry has thus far failed to identify a single neurobiological phenotypic marker or gene that is useful in making a diagnosis of a major psychiatric disorder or for predicting response to psychopharmacologic treatment.” “Epidemiological and clinical studies have shown extremely high rates of comorbidity among the disorders, undermining the hypothesis that the DSM syndromes represent distinct etiologies.” “The efficacy of medications cut across the DSM-defined categories,” regarding the same medications that are effective for depression and anxiety.
Most intimidating for the future of psychiatry is this last quotation stating that “reification of DSM-IV entities, to the point that they are considered equivalent to disease, is more likely to obscure than to elucidate research findings.”
A critical challenge for psychiatry involves the question whether it can go beyond the descriptive diagnosis.
Citing the abstract of the paper published by Steven Hyman titled ‘Can neuroscience be integrated into the DSM-V?’ (Nat Rev Neuroscience 2007 Sep;8(9):725-32.) We are confronted with the question of whether it is timely to ask if neuroscience has progressed to the point that the next editions of psychiatric diagnosis can usefully incorporate information about brain structure and function.
Simply phrased, we can ask is it time to go beyond descriptive psychiatry and toward a brain-related diagnosis of mental disorders.
Most opinion leaders would argue that it is too early, and we do not know enough for such a leap in psychiatric diagnosis.
Nevertheless I want to argue that by introducing computational-neuroscience and insights from neural-network science we have the tools to conceive a brain-related etiological diagnosis for mental disorders.
Neural-network models provide us with the theoretical mathematical and computational insights to the workings of the brain. They involve intuitions into how complex ensembles of units and processors represent and manipulate information, how they can form memory, recognition, logic, and decision making. They begin to explain awareness and consciousness, and potentially will be able to explicate emotions and feelings.
Neural network models are composed of units, each is a processor summing weighted inputs and outputting a threshold-dependent function. Connectivity among such active dynamic processors gives raise to self-organizing ensembles, and activation patterns, that support brain-like computational functions explained above.
The psychiatric literature has recently accumulated some studies where such models were used to explain mental disorders, one example involves the ‘disconnection syndrome’ hypothesis for psychosis and schizophrenia. This hypothesis argues that different brain processors become disconnected, statistically independent, fragmenting coherent brain organization resulting in the disintegration of experience with its typical psychotic manifestations.
Parenthetically this is not a totally new idea Theodor Meynert (teacher of Sigmund Freud) pioneered these ideas as early as the middle of the 18th century.
If we take this direction (of introducing computational-neuroscience to psychiatry) further, we can devise a comprehensive brain-related etiological diagnosis for psychiatry.
'Clinical Brain profiling' (CBP) is a translation of clinical phenomena to their related brain disturbance. It is based on insights from computational neuroscience, as such it proposes to go beyond the descriptive level of current psychiatry.
CBP has already been developed as a computer program; the program is actually a theory-based translation device where the input is translated to an output.
The input involves the set of clinical findings (currently practiced and also validate in comparison to accepted clinical scales), and the output is a set of brain-disturbances formulated as perturbation to optimal brain organization, thus for example perturbation to connectivity balance (among brain processors) or perturbation to hierarchical top-down bottom-up balance (and other detailed perturbations).
At this stage the output is inevitably hypothetical; nevertheless it is formulated in a testable-prediction manner. It can be proven or refuted based on imaging analysis, for example if the patient is presumably suffering from a connectivity breakdown having a 'disconnection' pattern of activity then applying ICA (Independent component analysis) to his fMRI or EEG will detect – and thus diagnose – his underlying brain pathology.
So the next step is to verify the theory (prove refute it, or update and develop it based on the empirical findings),
Once the descriptive part of CBP is found to be as-good-as the existing descriptive system, then a preliminary advantage of CBP can be a conceptual one, - in the sense that it will introduce a neuroscience-based terminology to psychiatry, - such terminology has immediate prelude advantages, first it will cause psychiatrist to talk neuroscience, second it will direct researchers toward brain-based pathology of mental disorders, and it can even have the benefit of being less stigmatizing, as the patient is not 'crazy' but has some 'mental coordination' problem like dyslexia or other more acceptable disturbance.
More importantly having brain-related orientation new therapeutic ideas can develop for more effective cures. For example, a disconnected brain can be 'reconnected' and 'optimized' using a brain-pacemaker (this is off course a simplified formulation); this is not a far-fetched idea, technology for controlling brain activity is being developed (e.g. TMS) the algorithm of intervention is the missing part, and CBP offers to provide it.
Experience-control using environmental enhancement devices such as virtual-reality, is another way in which brain organization can be optimized by experience -dependent-plasticity.
Two books are recommended for in-depth detailed explanation of the future diagnosis and related revolution in psychiatry.
‘Neuroanalysis’ by Psychology Press
‘Optimizers 2050’ Saga Books