Different brain systems support learning from received and avoided pain during human pain-avoidance learning

Area of Science:

• Neuroscience
• Computational Psychiatry
• Behavioral Economics

Background:

• Pain avoidance learning is crucial for survival.
• The neural and neurochemical underpinnings of learning from pain versus pain absence are not fully understood.
• Investigating shared or separate systems for processing pain and pain absence is essential.

Purpose of the Study:

• To determine if pain and pain absence avoidance learning utilize shared or distinct neural and neurochemical systems.
• To investigate the roles of the dopaminergic and opioidergic systems in pain-avoidance learning.
• To differentiate the neural encoding of prediction errors related to pain and no-pain outcomes.

Main Methods:

• Instrumental pain-avoidance learning task.
• Computational modeling to analyze learning rates.
• Functional magnetic resonance imaging (fMRI) and pharmacological manipulations (levodopa, naltrexone) in 83 participants.

Main Results:

• Computational modeling indicated participants learned more from received pain than avoided pain without intervention.
• Dopamine and opioid manipulations normalized learning asymmetry by increasing learning rates for avoided pain.
• fMRI showed distinct brain regions for pain prediction errors (subcortical/limbic) and no-pain prediction errors (cortical).
• Pharmacological manipulations did not alter the neural encoding of prediction errors.

Conclusions:

• Human pain-avoidance learning involves separate threat-learning (pain) and safety-learning (no-pain) systems.
• Dopamine and endogenous opioids specifically modulate learning from successfully avoided pain.
• Distinct neural pathways encode prediction errors for received versus avoided pain outcomes.