What is simulated in the action observation network when we observe actions?

Area of Science:

• Cognitive Neuroscience
• Neuroscience
• Psychology

Background:

• Extensive research investigates action simulation for inferring intentions.
• Mirror neurons and the action observation network (AON) are key areas of focus.
• The precise nature of simulation and its role in intention inference remain unclear.

Purpose of the Study:

• To clarify what is simulated during action observation.
• To understand how simulations enable intention inference, particularly with varying kinematics.
• To investigate if the action observation system (AON) can infer the same goal despite different movement patterns (e.g., left vs. right hand grasping).

Main Methods:

• Conducted a behavioral study with healthy human subjects.
• Assessed subjects' ability to detect subtle kinematic changes in observed actions.
• Fitted behavioral responses using a computational model based on the predictive coding account of mirror neurons.

Main Results:

• Subjects demonstrated high sensitivity to subtle variations in observed action kinematics.
• The predictive coding model successfully accounted for all observed behavioral effects.
• This supports a Bayesian framework for action observation, grounded in the free-energy principle.

Conclusions:

• The action observation network (AON) likely operates within a predictive coding framework.
• This framework can explain how the brain infers intentions from observed actions, even with kinematic variability.
• Simulation in the AON, viewed through predictive coding, enables robust goal inference.