Neural coding of prior expectations in hierarchical intention inference

Area of Science:

• Neuroscience
• Cognitive Science
• Social Cognition

Background:

• Inferring intentions is vital for social interaction.
• This inference involves integrating sensory evidence with prior expectations.
• The abstractness of an intention influences this integration process.

Purpose of the Study:

• Investigate neural mechanisms that adjust evidence integration based on intention abstractness.
• Examine how the brain balances prior knowledge and sensory input for intention inference.

Main Methods:

• Utilized functional magnetic resonance imaging (fMRI) in two experiments.
• Presented participants with action scenes depicting varying intention abstractness (Superordinate vs. Basic, Social vs. Non-social).
• Manipulated prior expectations and sensory evidence levels.

Main Results:

• Participants relied more on prior expectations when sensory evidence was limited.
• Medial prefrontal cortex (mPFC) activity reflected the interplay between prior and sensory information.
• Increased abstractness of intentions (Superordinate > Basic, Social > Non-Social) enhanced backward connectivity between mPFC and temporo-parietal junction (TPJ).

Conclusions:

• The fronto-parietal network plays a key role in modulating the influence of prior beliefs versus sensory data.
• This network adjusts the weight of evidence based on the hierarchical abstractness of inferred intentions.
• Findings elucidate neural underpinnings of adaptive social cognition and intention understanding.