Models of the posterior parietal cortex which perform multimodal integration and represent space in several...

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Neurons in the posterior-parietal cortex (PPC) exhibit saccadic responses to visual and auditory targets.
• These responses are modulated by eye and head position, indicating multisensory integration.
• PPC may encode saccadic targets in various coordinate frames, including eye-centered, head-centered, and body-centered.

Purpose of the Study:

• To investigate coordinate transformations within the PPC.
• To compare computational models of PPC with different output representations (eye-centered, head-centered, body-centered).
• To determine how gain fields (GFs) contribute to coordinate transformations.

Main Methods:

• Developed and trained computational models of PPC with varying output layer representations.
• Analyzed the properties of hidden layer units, including gain fields and coding strategies.
• Compared model unit properties to known characteristics of PPC neurons.

Main Results:

• Models with multiple output representations (eye, head, body-centered) better matched PPC neuron properties.
• Hidden layer units exhibited gain fields for eye and/or head position, depending on the model's output.
• Intermediate coding, where units represent targets in mixed coordinate frames, was observed.
• Coordinate transformations were achieved through specific relationships between receptive fields and gain fields.

Conclusions:

• Gain fields are effective mechanisms for performing coordinate transformations in neural circuits.
• The PPC likely utilizes multiple coordinate frames for representing spatial targets.
• Neurons with intermediate coding are expected during the process of coordinate transformation in the PPC.