Komputationale Modelle in der psychiatrischen Forschung
Abstract
Zusammenfassung. Komputationale Methoden werden in einem weiten Bereich psychiatrischer Forschung eingesetzt v.a. beim Umgang mit großen Datensätzen. Hier diskutieren wir, wie mit Hilfe komputationaler Methoden das Entscheidungsverhalten in bestimmten Testsituationen in mathematischen Gleichungen abgebildet werden kann, um diesem Entscheidungsverhalten wiederum neurobiologischen Korrelaten zuordnen zu können. Ein bekanntes Beispiel ist das belohnungsabhängige Lernen, bei dem die Größe eines Vorhersagefehlers, d.h. des Unterschieds zwischen der erwarteten und der eingetroffenen Belohnung, mathematisch berechenbar ist und mit entsprechenden Signalen wie der Dopaminfreisetzung im ventralen Striatum in Tiermodellen oder entsprechenden funktionellen Bildgebungskorrelaten beim Menschen in Verbindung gebracht werden kann. Veränderungen dieser Lernmechanismen lassen sich nosologieübergreifend bei verschiedenen Störungsbildungen wie Suchterkrankungen, Psychosen und majoren affektiven Erkrankungen nachweisen und ermöglichen so einen dimensionalen Ansatz, mit dem neurobiologische Korrelate einzelner Mechanismen in verschiedenen Krankheitsbildern identifiziert und mit Teilaspekten der Symptomatik in Verbindung gebracht werden können. Die vorliegende Arbeit diskutiert entsprechende Ansätze im Bereich Pawlowschen Konditionierens und des belohnungsabhängigen, zielgerichteten und habituellen Entscheidungsverhaltens. Der Fokus auf Lernmechanismen betont hierbei die Vielfältigkeit und Veränderbarkeit menschlichen Verhaltens, beispielsweise durch Trainingsprogramme oder kognitiv-behaviorale Interventionen.
Abstract. Computational methods are widely used in psychiatric research and comprise big data approaches as well as the here discussed research approach, that mathematically models decision making in order to identify its neurobiological correlates. A well-known example is reward-dependent learning, where the prediction error (the difference between expected and actual reward) is mathematically computable and can be linked to neural signals such as dopamine release in the ventral striatum of animals and its functional correlates in humans. Changes in relevant learning mechanisms can be detected across different mental disorders including substance use disorders, psychoses and major depressive disorders thus providing a dimensional approach to identify neurobiological correlates of specific learning-mechanisms and their associated symptoms in different mental disorders. Here we discuss key approaches in the area of Pavlovian conditioning as well as reward-dependent (goal-directed and habitual) decision making. The focus on learning mechanisms emphasizes the diversity and modifiability of human behavior, which can be targeted therapeutically with training programs and cognitive-behavioral interventions.
Literatur
(2016). Computational Psychiatry: towards a mathematically informed understanding of mental illness. Journal of Neurology, Neurosurgery & Psychiatry , 87 , 53–63.
(2013). The computational anatomy of psychosis. Frontiers in Psychiatry , 4 , 47.
(1986). Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annual Review of Neuroscience , 9 , 357–381.
(2010). Effect of brain structure, brain function, and brain connectivity on relapse in alcohol-dependent patients. Archives of General Psychiatry , 69 , 842–852.
(2010). Altered representation of expected value in the orbitofrontal cortex in mania. Human Brain Mapping , 31 , 958–969.
(2016). Psychosen: Ringen um Selbstverständlichkeit . Köln: Psychiatrie Verlag.
(2011). Model-based influences on humans’ choices and striatal prediction errors. Neuron , 69 , 1204–1215.
(2015). Chronic alcohol intake abolishes the relationship between dopamine synthesis capacity and learning signals in the ventral striatum. European Journal of Neuroscience , 41 , 477–486.
(2016). Pavlovian-to-instrumental transfer effects in the nucleus accumbens relate to relapse in alcohol dependence. Addiction Biology , 21 , 719–731.
(2014). Too difficult to stop: mechanisms facilitating relapse in alcohol dependence. Neuropsychobiology , 70 , 103–110.
(2013). Pawlowsch-Instrumentelle Transfereffekte bei Alkoholabhängigkeit. SUCHT , 59 , 215–223.
(2004). Cue-induced activation of the striatum and medial prefrontal cortex is associated with subsequent relapse in abstinent alcoholics. Psychopharmacology , 175 , 296–302.
(2000). Striatonigrostriatal pathways in primates form an ascending spiral from the shell to the dorsolateral striatum. The Journal of Neuroscience , 20 , 2369–2382.
(2015). Dimensional psychiatry: reward dysfunction and depressive mood across psychiatric disorders. Psychopharmacology (Berl) , 232 , 331–341.
(2000). Das dopaminerge Verstärkungssystem. Funktion, Interaktion mit anderen Neurotransmittersystemen und psychopathologische Korrelate. Darmstadt: Steinkopff.
(1998). Psychomotor slowing, negative symptoms and dopamine receptor availability – an IBZM SPECT study in neuroleptic-treated and drug-free schizophrenic patients. Schizophrenia Research , 31 , 19–26.
(2003). Reward craving and withdrawal relief craving: assessment of different motivational pathways to alcohol intake. Alcohol and Alcoholism , 38 , 35–39.
(2010). Dopaminergic dysfunction in schizophrenia: salience attribution revisited. Schizophrenia Bulletin , 36 , 472–485.
(2016). The specificity of Pavlovian regulation is associated with recovery from depression. Psychological Medicine , 46 , 1027–1035.
(2012). Ventral striatal activation during reward processing in subjects with ultra-high risk for schizophrenia. Neuropsychobiology , 66 , 50–56.
(2001). Anticipation of increasing monetary reward selectively recruits nucleus accumbens. Journal of Neuroscience , 21 , RC159.
(2008). Abnormal temporal difference reward-learning signals in major depression. Brain , 131 (Pt 8), 2084–2093.
(2011). A bayesian foundation for individual learning under uncertainty. Frontiers in Human Neuroscience , 5 , 39.
(2003). Temporal difference models and reward-related learning in the human brain. Neuron , 38 , 329–337.
(2010). Prefrontal cortex fails to learn from reward prediction errors in alcohol dependence. Journal of Neuroscience , 30 , 7749–7753.
(1927). Conditioned reflexes: an investigation of the pysiological activity of the cerebral cortex. Oxford: Oxford University Press.
(1993). The neural basis of drug craving: An incentive-sensitization theory of addiction. Brain Research Reviews , 18 , 247–291.
(2014). Striatal dysfunction during reversal learning in unmedicated schizophrenia patients. Neuroimage , 89 , 171–180.
(1997). A neural substrate of prediction and reward. Science , 275 (5306), 1593–1599.
(2014). Model-based and model-free decisions in alcohol dependence. Neuropsychobiology , 70 , 122–131.
(2016). Don't think, just feel the music: individuals with strong pavlovian-to-instrumental transfer effects rely less on model-based reinforcement learning. Journal of Cognitive Neuroscience , 28 , 985–995.
(2014). Computational approaches to psychiatry. Current Opinions in Neurobiology , 25 , 85–92.
(2015). Disorders of compulsivity: a common bias towards learning habits. Molecular Psychiatry , 20 , 345–352.
(2014). Neural correlates of alcohol-approach bias in alcohol addiction: the spirit is willing but the flesh is weak for spirits. Neuropsychopharmacology , 39 , 688–697.
(2011). Retraining automatic action tendencies changes alcoholic patients' approach bias for alcohol and improves treatment outcome. Psychological Science , 22 , 490–497.
(2007). Dysfunction of reward processing correlates with alcohol craving in detoxified alcoholics. Neuroimage , 35 , 787–794.
