Associability-modulated loss learning is increased in posttraumatic stress disorder.

Area of Science:

• Neuroscience
• Computational Psychiatry
• Psychology

Background:

• Posttraumatic stress disorder (PTSD) is characterized by disproportionate reactions to environmental stimuli.
• Heightened responses in PTSD may stem from disruptions in reinforcement learning processes.
• Anomalous attention gating during learning (associability) is hypothesized to underlie overreactions in PTSD.

Purpose of the Study:

• To investigate the association between heightened responses in PTSD and specific components of reinforcement learning.
• To test the mechanistic hypothesis that overreactions in PTSD result from aberrant attention gating (associability) during learning.
• To identify potential neuromechanistic targets for PTSD treatment.

Main Methods:

• Employed functional neuroimaging (fMRI) in combat-deployed veterans with and without PTSD.
• Utilized a loss-learning task combined with a computational reinforcement learning model.
• Analyzed trial-by-trial prediction errors and associability values, alongside neural correlates.

Main Results:

• Neural substrates of associability value and behavioral estimates of associability updating increased with PTSD severity during loss learning.
• Prediction error components did not significantly increase with PTSD.
• The interaction between PTSD severity and neural associability markers predicted behavioral choices.

Conclusions:

• Increased attention-based learning (associability) may underlie specific aspects of posttraumatic stress disorder.
• Findings suggest that aberrant associability, not prediction error, is linked to PTSD symptomology.
• These results highlight potential neuromechanistic treatment targets for PTSD focused on attention regulation.