On two very interesting meetings in Vienna 2007

Scroll down to see some pictures from the meetings

At the two days Neuropsychoanalysis congress titled “neuro-psychoanalytic perspectives on depression”

Hugo Bleichmar argues the case for the need to develop an integrative clinical psychoanalytic model of depression which takes into account various pathways leading to different subtypes of depression. Aggression, guilt, narcissistic disorders, persecutory anxieties, ego deficits, masochism, identification with depressive parents and fixation to traumatic events in which the subject felt helpless are factors intervening in the genesis and maintenance of depression. Relationship between aggression and depression exist and aggression can turn into depression. Guilt can be independent of aggression. Psychoanalytic interventions that could be pertinent for a given subtype of depression, but which could prove counterproductive when applied to another subtype were discussed.

Helen Mayberg from the Institute of Medical Science, University of Toronto, Canada talked about her research of Deep Brain Stimulation (DBS) for treatment-resistant Depression: the DBS stimulator is implanted in the subgenual cingulate region shows prompt promising antidepressant effect. Helen described patients’ reactions and feelings when the DBS was turned on and off.

Jaak Panksepp from the Department of Veterinary Comparative Anatomy, Physiology, and Pharmacology, Washington State University, College of Veterinary Medicine, claimed that by studying and understanding emotions at the neural level we can understand emotions and emotional disorders in humans. He calls this field of science ‘affective neuroscience.’ He talked about organization of emotions at the brain level, anticipatory/expectancy brain mechanisms, social-emotional mechanisms in the brain, play/joy processes in the brain, separation, anxiety, and fear organization in the brain. He showed that play in rat pups increases considerably after the socially depriving the animals for a period of time. Play behavior in rat pups, classified as "rough and tumble", involves pinning, chasing, and rolling, as well as an element of deception and surprise attack. Panksepp argues that the rat pups do not become aggressive when playing. He argues that playing serves a deeper function than simple recreation. He contends play factors in to optimal brain development. He claims that playing behavior releases opiods into the brain. Indeed, the frontal lobe of the brain grows as a response to playing behavior. Panksepp hypothesizes that the opiods released during play act to stimulate further play; eventually, the opiod level rises to a high enough level satisfy the need to play by inducing a feeling of "social comfort." Additionally according to Panksepp the early play and its related hormonal release maybe a protective factor against adult depression.  

Georg Northoff from the Department of Psychiatry, Psychotherapy and Psychosomatics, University of Magdeburg, Magdeburg, Germany, talked about investigating the psychodynamic-neuronal relationship, he thinks this consists in systematically linking the individual contents of first-person subjective experience to third-person observation of neuronal states. He argued that combined psychodynamic and imaging studies suggest that sensorimotor regression might be associated with dysfunction in the neural network including the orbitofrontal, the medial prefrontal and the premotor cortices. In general sensorimotor regression and other defense mechanisms are psychoanalytic constructs that are hypothesized to be complex emotional-cognitive constellations. He contested Merker argument that subcortical regions are sufficient for the constitution of consciousness as "immediate, unreflective experience" as distinguished from self-consciousness. He thinks that Merker neglects the differentiation between pre-reflective self-awareness and reflective self-consciousness. Pre-reflective self-awareness allows us to immediately and unreflectively experience our self, which functionally may be mediated by what he calls self-related processing in subcortical regions.

David Taylor from Psychotherapy Evaluation Research Unit, Tavistock Centre, London, UK, talked about treatment-resistant depression, using a systematic search of studies evaluating a psychological intervention in adults diagnosed for major depressive disorder. Psychological treatments for depression are commonly delivered and often recommended following the failure of medication. The rareness of evidence for their effectiveness in these situations is a significant problem. There is a need for studies with a strong controlled design investigating the effectiveness of psychological treatments for patients with treatment-resistant depression.

My impression from the meeting (which I voiced in the discussion) is that in order to really bridge neuroscience and psychoanalysis neural computation insights must be introduced. Otherwise the congress seems to equate a regular neuroscience meeting with metaphorical remarks comparing neurological findings with psychoanalytic concepts. Neuralcomputation is a ‘hard tool’ to really bridge psychoanalytic phenomena and neuroscience. For example the concept of ‘transference’ relates to activation of neural ensembles representing past experience (with relevant other .i.e., object) instead of current experience (real-time input) this can easily be modeled with a highly connected Hebbian-trained Hopfield neural-network. The mathematics that these networks provide may one day allow for signal-processing-based identification of transference in the brain.        

ENF2007,

Immediately after the Neuropsychoanalysis meeting there was a one-day ENF 2007 meeting conceived and organized by Dietmar Dietrich from the computer and electrical engineering faculty of the Vienna Institute of Technology.

This was a really unique meeting bringing engineers of artificial intelligence (AI) and Neuropsychoanalysts together to discuss if emotions can be implemented in AI machines. This off course entails ‘Emulating the brain’ which was the title of the meeting.

Etienne Bernard gave an overview, written together with Brigitte Lorenz, about AI focusing on computational models of emotions. Then Dietrich talked about the technical realization of a neuro-psychoanalytic model of the mind, the constraints and vision when trying to map Sigmund Freud’s model and Luria’s dynamic neuropsychology into a machine. The talk involved models of visual integration and language concept representations and an outlook on layered structure architectures that would be required for future developments. Mark Solms representing (and heading) the Neuropsychoanalysts talked about the psychoanalytic stand point being subjective and that brain events in comparison are different observational perspectives. He described the ‘mind’ in terms of ‘consciousness’ ‘intentionality’ and ‘agency’ working for the body to maximize the chances of the body to survive and to reproduce. Aaron Sloman from school of computer science in Birmingham the UK, gave an invited talk about natural minds being information-processing ‘virtual’ machines produced by evolution. Affective states including emotions can be construed as aspects of the ‘control’ mechanisms of minds. Jaak Panksepp emphasized that the only experiencing minds that exists are those linked to the dynamics of leaving bodies. Later discussions debated the need of bodily simulations for AI development and went further to discuss science-fiction possibilities about robots feelings and relations toward man and among themselves.

My comments which I voiced privately in the dinner discussion with Aaron Sloman included the need for brain models to provide simulations of mental disorders. The models should emulate the brain to the extent that when psychopathology is simulated it can generate a testable prediction for finding the neural basis of mental disorders in real brains of real patients. Actually this is what ‘NeuroAnalysis’ is all about. According to Aaron we are still far away from that.