Frontostriatal dynamics modelling obsessions and compulsions
Poster No:
1078
Submission Type:
Abstract Submission
Authors:
Sebastien Naze1, Luke Hearne1, Paula Sanz-Leon2, Conor Robinson1, Caitlin Hall1, Saurabh Sonkusare3, Bjorn Burgher1, Andrew Zalesky4, James Roberts1, Luca Cocchi1
Institutions:
1QIMR Berghofer Medical Research Institute, Brisbane, Queensland, Australia, 2School of Physics, Faculty of Science, The University of Sydney, Sydney, New South Wales, Australia, 3University of Newcastle, Newcastle, New South Wales, Australia, 4Systems Lab, Department of Psychiatry, The University of Melbourne, Melbourne, Victoria, Australia
First Author:
Co-Author(s):
Paula Sanz-Leon
School of Physics, Faculty of Science, The University of Sydney
Sydney, New South Wales, Australia
School of Physics, Faculty of Science, The University of Sydney
Sydney, New South Wales, Australia
Andrew Zalesky
Systems Lab, Department of Psychiatry, The University of Melbourne
Melbourne, Victoria, Australia
Systems Lab, Department of Psychiatry, The University of Melbourne
Melbourne, Victoria, Australia
Introduction:
The activity of the frontostriatal system supports individuals' thought processes and behavioural patterns that can become maladaptive in obsessive-compulsive disorder (OCD). Converging evidence from preclinical and clinical work suggests that OCD maps onto a functional imbalance in the ventral and dorsal frontostriatal circuits. However, the neural mechanisms supporting these dysregulations remain elusive, their association with symptom severity is unclear, and therapeutic interventions are limited. Progress towards a better understanding of frontostriatal deregulations in OCD is limited by the difficulty in assessing in-vivo changes in neural mechanisms supporting frontostriatal function in humans. To address these gaps, we combine neuroimaging and behavioural data from individuals with OCD and controls with advanced computational modelling.
Methods:
Longitudinal functional MRI and clinical measures of symptoms' manifestation from 52 subjects with OCD and 45 healthy controls were collected in the context of a clinical trial (ACTRN12616001687482). We developed a computational model incorporating dynamical properties relevant to goal-directed behaviours and decision-making through coupled bistable systems governing cortico-striato-thalamo-cortical (CSTC) activity. We use a sequential Bayesian optimisation algorithm to infer CSTC features that empirically differentiate healthy from pathological systems' parameters. A combinatorial approach is then used to predict clinical outcomes from virtual interventions on the model, operated via systematic analysis of parameter permutations from OCD to healthy posterior distributions. The effects of restoring distinct parameter(s) are quantified based on the resulting improvements in functional connectivity and statistical differentiation with respect to intervention-free cohorts. Next, we tested the validity of the model predictions using a digital twin paradigm whereby longitudinal data from subjects with OCD are paired with virtual subjects through a distance metric in functional connectivity space. The approach is illustrated in Figure 1.
Results:
We find that multiple parameters in the model show significant difference between OCD and healthy controls (Figure 2). The virtual intervention analysis reveals that bidirectionally decreasing spontaneous neural coupling in the ventromedial (affective) circuit while concurrently increasing dorsolateral (cognitive) cortico-striatal coupling delivers the highest functional improvements in subjects with OCD. The analysis of longitudinal changes in obsessions and compulsions with respect to modelled neural interventions directly supports those predictions. Importantly, the analysis of virtual interventions suggests that targeting the ventromedial circuit alone, despite being the most dysregulated, would not deliver the best functional outcomes unless there is concomitant targeting of the dorsolateral circuit.
Conclusions:
By highlighting behaviourally meaningful neural mechanisms hidden from traditional neuroimaging analysis, this study advances knowledge on the neural basis of OCD. The study also provides new therapeutic targets to treat maladaptive obsessions and compulsions.
Brain Stimulation:
Non-invasive Magnetic/TMS
Sonic/Ultrasound
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 2
Higher Cognitive Functions:
Executive Function, Cognitive Control and Decision Making
Modeling and Analysis Methods:
Bayesian Modeling 1
Keywords:
Basal Ganglia
Computational Neuroscience
Data analysis
FUNCTIONAL MRI
Informatics
Modeling
Obessive Compulsive Disorder
Systems
Transcranial Magnetic Stimulation (TMS)
1|2Indicates the priority used for review