Prior expectation enhances sensorimotor behavior by modulating population tuning and subspace activity in sensory...

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Prior knowledge significantly enhances perception and goal-directed behaviors, especially with limited or noisy sensory input.
• The precise neural mechanisms by which prior expectations improve sensorimotor functions remain largely unexplored.

Purpose of the Study:

• To investigate the neural basis of how prior expectations enhance sensorimotor behavior.
• To examine neural activity in the middle temporal (MT) area during a smooth pursuit eye movement task guided by directional expectations.

Main Methods:

• Recorded neural activity in the MT area of the visual cortex in monkeys performing a smooth pursuit eye movement task.
• Manipulated prior expectations of the visual target's motion direction.
• Analyzed neural population responses and simulated their impact on behavior using a realistic MT model.
• Employed state-space analysis to identify neural signals related to prior expectations.

Main Results:

• Prior expectations selectively reduced MT neural responses based on preferred directions when sensory evidence was weak.
• This response modulation led to sharpened neural population direction tuning.
• Simulations indicated that sharpened tuning could account for observed biases and variabilities in smooth pursuit eye movements.
• State-space analysis revealed neural signals in MT activity that correlated with behavioral changes due to prior expectations.

Conclusions:

• Neural computations within the sensory area (MT) alone can integrate prior knowledge and sensory evidence.
• Sharpened neural population tuning in the MT area is a key mechanism underlying the behavioral benefits of prior expectations.
• Prior expectations are neurally encoded in MT population activity, influencing sensorimotor control.