Force tuning explains changes in phasic dopamine signaling during stimulus-reward learning.

Area of Science:

• Neuroscience
• Behavioral Science
• Computational Neuroscience

Background:

• The prevailing hypothesis suggests phasic dopamine (DA) activity encodes reward prediction errors (RPEs) crucial for reinforcement learning.
• Recent studies indicate DA neurons are vital for performance, not learning itself.
• Limitations in previous research stem from restricted behavioral measures.

Approach:

• This study recorded and manipulated DA activity in the ventral tegmental area (VTA) during stimulus-reward learning.
• Subtle force exertion was measured as a key behavioral output.
• Investigated the role of VTA DA neurons in motivated behavior.

Key Points:

• Two main populations of DA neurons were identified, increasing firing before forward and backward force exertion.
• Force tuning remained consistent irrespective of learning, reward predictability, or outcome valence.
• Observed changes in force exertion patterns explained traditional RPE-supporting findings, including reward magnitude, probability, and unpredicted outcomes.

Conclusions:

• Ventral tegmental area (VTA) dopamine (DA) neurons do not signal reward prediction errors (RPEs).
• These DA neurons play a critical role in regulating force exertion during motivated behaviors.
• The findings challenge the RPE hypothesis and propose a new function for VTA DA signaling.