Area MT carries acceleration information in a quickly and directly decodable representation.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Visual Perception

Background:

• Neurons in area MT are known to encode visual velocity.
• Estimating motion acceleration typically involves calculating the rate of change of velocity.
• The brain may use indirect or direct mechanisms to represent motion acceleration.

Purpose of the Study:

• To investigate the neural representation of visual motion acceleration.
• To compare indirect velocity-based and direct neural response-based decoding of acceleration.
• To examine acceleration processing in areas MT and MST.

Main Methods:

• Recorded neural activity from ensembles of MT neurons in awake, fixating macaques.
• Presented stimuli featuring linearly accelerating visual motion.
• Analyzed neural responses to decode motion acceleration using direct and indirect methods.
• Performed a similar analysis on activity in the medial superior temporal area (MST).

Main Results:

• Direct decoding of motion acceleration from MT neurons was feasible on faster timescales and with higher fidelity than indirect decoding.
• Motion acceleration information was efficiently extracted from heterogeneous MT ensemble responses.
• Area MST did not show a more refined representation of motion acceleration compared to MT.

Conclusions:

• The brain can efficiently extract motion acceleration information through direct decoding of MT neural responses, exploiting nonlinearities.
• This direct mechanism operates faster and more accurately than indirect velocity calculation.
• Neural systems may opportunistically leverage response idiosyncrasies for efficient information extraction.