Effective connectivity during animacy perception--dynamic causal modelling of Human Connectome Project data

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Neuroscience

Background:

• Biological systems are complex to model and predict.
• Predictive coding involves comparing predicted sensory input with actual feedback, passing prediction errors to higher brain areas.
• Animate motion, compared to inanimate motion, is hypothesized to generate greater prediction errors, influencing neural signal flow.

Purpose of the Study:

• To investigate effective neural connectivity differences between processing animate and inanimate motion.
• To test the hypothesis that animate motion increases signal passing from visual area V5 to posterior superior temporal sulcus (pSTS).

Main Methods:

• Utilized a large-scale fMRI dataset from the Human Connectome Project.
• Analyzed brain activity in 132 participants viewing animations of triangles with either intentional (animate) or mechanical (inanimate) motion.
• Examined effective connectivity, specifically focusing on the pathway from V5 to pSTS.

Main Results:

• Forward connectivity from visual area V5 to the posterior superior temporal sulcus (pSTS) significantly increased when viewing animate motion.
• Inhibitory self-connection within the pSTS decreased during the perception of animate motion compared to inanimate motion.
• These findings support the predictive coding model for motion perception.

Conclusions:

• Animate motion elicits prediction errors that modulate neural connectivity between visual processing areas.
• Increased attention to animate stimuli may result from these prediction errors.
• The study provides insights into the neural mechanisms underlying the perception of intentional movement.