Structural synaptogenesis superior to functional modulation in a pruning-based recurrent network model of OCD

Area of Science:

• Computational neuroscience
• Neurobiology of mental disorders
• Cognitive science

Background:

• Obsessive-compulsive disorder (OCD) involves intrusive thoughts and repetitive behaviors, with current treatments showing limitations.
• Existing models focusing on neurotransmitters may not fully explain OCD.
• Abnormal synaptic pruning is a potential mechanism contributing to cortico-striato-thalamo-cortical circuit rigidity in OCD.

Purpose of the Study:

• To investigate the role of excessive synaptic pruning in cognitive inflexibility, a key feature of OCD.
• To simulate and compare the effects of different intervention mechanisms on a computational model of cortico-striato-thalamo-cortical dynamics.

Main Methods:

• Developed a recurrent neural network model simulating cortico-striato-thalamo-cortical dynamics.
• Implemented activity-dependent synaptic pruning to mimic excessive pruning.
• Simulated three interventions: ketamine-like structural repair, SSRI-like adaptation, and neurosteroid-like inhibition.

Main Results:

• The untreated pruned network exhibited impaired accuracy and elevated perseveration.
• Ketamine-like repair demonstrated the most robust reduction in perseveration.
• SSRI-like adaptation showed higher efficiency but greater relapse vulnerability compared to neurosteroid-like inhibition.

Conclusions:

• Excessive synaptic pruning is a plausible contributor to OCD-like cognitive inflexibility.
• Structural and functional interventions present different trade-offs in computational models.
• Findings are hypothesis-generating and do not provide clinical treatment recommendations.