Anxiety, avoidance, and sequential evaluation.

Area of Science:

• Computational psychiatry
• Cognitive neuroscience
• Behavioral economics

Background:

• Anxiety disorders involve aberrant threat processing, but core pathogenesis remains unclear.
• Existing models often overlook the role of specific belief biases in anxiety.
• Maladaptive avoidance is a key feature of anxiety disorders.

Purpose of the Study:

• To propose and formalize a computational model where pessimistic beliefs drive anxiety symptoms.
• To explain how a specific belief bias can account for diverse anxiety phenomena.
• To link computational insights to existing psychiatric theories of anxiety.

Main Methods:

• Decision theoretic analysis of avoidance behavior.
• Reinforcement learning model incorporating belief biases.
• Simulations reproducing laboratory findings in anxiety.

Main Results:

• A pessimistic bias regarding future avoidance leads to excessive fear and avoidance propagation.
• The model explains exaggerated threat appraisals, fear generalization, and persistent avoidance.
• Simulations match empirical data on abnormal decision-making in anxiety.

Conclusions:

• Unrealistically pessimistic assumptions about avoidance can be a core mechanism in anxiety disorders.
• This single bias unifies disparate anxiety symptoms and links to depression.
• The model offers a formalization of maladaptive beliefs about control and self-efficacy.