Causal Inference in Audiovisual Perception.

Area of Science:

• Neuroscience
• Cognitive Science
• Perception

Background:

• The brain continuously processes multisensory signals to construct a coherent perception of the environment.
• A key challenge is determining whether signals originate from a common cause (integration) or separate causes (segregation).
• The neural mechanisms underlying this causal inference problem remain incompletely understood.

Purpose of the Study:

• To investigate the neural basis of causal inference in multisensory integration.
• To identify brain regions involved in determining the causal structure of auditory and visual signals.
• To differentiate the roles of brain areas in inferring causal structure versus guiding motor responses.

Main Methods:

• Functional magnetic resonance imaging (fMRI) was employed in human participants.
• Participants performed a task requiring causal decisions about synchronous auditory and visual signals presented at varying spatial locations.
• Multivariate fMRI pattern analysis was used to decode neural representations of causal inference and spatial information.

Main Results:

• The lateral prefrontal cortex uniquely encoded the outcome of causal inference (common vs. separate causes).
• The frontal eye field (FEF) and intraparietal sulcus (IPS) circuitry concurrently processed spatial, decisional, and motor response information.
• Individual adjustments of audiovisual disparity ensured that causal decisions were based on inference rather than simple spatial cues.

Conclusions:

• The lateral prefrontal cortex is critical for inferring the causal structure of sensory inputs.
• The FEF-IPS circuitry integrates multisensory spatial information, guided by inferred causal structure, to inform motor actions.
• This study elucidates distinct neural pathways for causal inference and sensorimotor integration in a complex sensory environment.