Dual computational systems in the development and evolution of mammalian brains

Area of Science:

• Neuroscience
• Computational Neuroscience
• Evolutionary Biology

Background:

• Mammalian brain evolution exhibits consistent covariation between major components.
• A notable inverse relationship exists between the limbic system and the neocortex.

Purpose of the Study:

• To investigate the functional basis for the observed inverse relationship between brain components.
• To explore how evolutionary optimization shapes neural representations.

Main Methods:

• Mapped multidimensional representations of task-optimized artificial neural networks onto 2D surfaces.
• Simulated networks optimized for various sensory and cognitive tasks (visual, auditory, somatosensory, olfactory, relational memory).

Main Results:

• Networks for sensory tasks (visual, somatosensory, auditory) formed ordered, spatiotopic maps.
• Networks for olfactory and relational memory tasks developed fractured, distributed maps.
• Evolutionary optimization led to inverse covariation between ordered and disordered network components.

Conclusions:

• The pattern of brain component covariation reflects an essential computational duality.
• This duality arises from competition for representational space during evolutionary optimization.
• Suggests a fundamental principle governing brain evolution and function.