A Canonical Laminar Neocortical Circuit Whose Bottom-Up, Horizontal, and Top-Down Pathways Control Attention,...

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Neuroscience

Background:

• The human cerebral cortex is organized into distinct layers.
• These layers form a canonical laminar network with bottom-up, horizontal, and top-down pathways.
• This network supports diverse biological intelligence across neocortical areas.

Purpose of the Study:

• To describe the canonical laminar network of the human cerebral cortex.
• To explain how this network supports varied cognitive processes.
• To elucidate the mechanisms underlying interstream interactions and their role in goal-oriented behaviors.

Main Methods:

• Review and synthesis of existing models of cortical organization.
• Analysis of computational principles governing parallel cortical streams.
• Explanation of how interstream interactions overcome computational deficiencies.

Main Results:

• The canonical laminar network supports diverse functions including 3D vision, decision-making, speech perception, and working memory.
• Parallel cortical streams operate under computationally complementary laws.
• Interactions between these streams are crucial for effective goal-oriented behaviors.

Conclusions:

• The described canonical network provides a unified framework for understanding diverse cognitive functions.
• Interactions between parallel cortical streams are essential for integrating information and enabling complex behaviors.
• Models of this network explain the emergence of complementary properties and their role in overcoming computational limitations.