Computational Modelling Reveals Slower Safety Learning and Threat Extinction are Associated With Higher Anxiety...

Area of Science:

• Cognitive Neuroscience
• Clinical Psychology
• Computational Psychiatry

Background:

• Anxiety disorders involve exaggerated responses to threats and impaired threat discrimination and extinction.
• Fear conditioning models, including acquisition and extinction phases, are used to study these impairments.
• Traditional analysis methods have limitations in capturing trial-by-trial cognitive mechanisms.

Purpose of the Study:

• To investigate cognitive mechanisms underlying anxiety disorders using computational modeling of fear conditioning.
• To examine the relationship between learning rates (threat learning, safety learning, threat extinction) and anxiety severity.
• To identify individual differences in cognitive processes contributing to anxiety persistence.

Main Methods:

• A normative sample (n=145) completed a differential fear conditioning task via a smartphone app.
• Anxiety severity was assessed using the GAD-7 measure.
• Computational models were applied to task data to estimate trial-by-trial learning rates.

Main Results:

• Threat learning rate did not correlate with anxiety severity.
• Threat extinction and safety learning rates showed small negative associations with anxiety severity (ρ = -0.22, p = 0.01 & ρ = -0.21, p = 0.01).
• These findings suggest anxious individuals learn safety cues more slowly.

Conclusions:

• Impaired fear extinction is a significant factor in anxiety disorders.
• Learning rates, particularly for safety and extinction, serve as important indices of anxiety severity.
• Computational modeling offers a powerful tool to uncover hidden cognitive mechanisms in anxiety.