Rarely categorical, always high-dimensional: how the neural code changes along the cortical hierarchy

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• The organization of neural populations into functionally distinct groups for information encoding remains a central debate in neuroscience.
• Understanding whether neural representations are "categorical" or continuous is crucial for deciphering neural computation.

Purpose of the Study:

• To systematically analyze how cortical neurons encode cognitive, sensory, and movement variables across multiple brain regions.
• To investigate the scale-dependent nature of neural coding and its variation across the cortical hierarchy.
• To explore the relationship between neural selectivity, representation dimensionality, and computational flexibility.

Main Methods:

• Analysis of over 14,000 neuronal units from the International Brain Laboratory's public Brain-wide Map dataset.
• Systematic examination of neural encoding across 43 cortical regions during a complex task.
• Mathematical modeling to explain the link between single-neuron selectivity and population code properties.

Main Results:

• Neural coding structure is scale-dependent: categorical on a whole-cortex scale, but non-categorical within most individual regions.
• Categorical representations were primarily observed in primary sensory areas.
• A strong inverse correlation was found between categorical clustering and the dimensionality of neural representations.
• The fraction of linearly separable conditions was near maximal across all cortical areas, indicating robust encoding flexibility.

Conclusions:

• Cortical neural codes are predominantly non-categorical and high-dimensional, except in lower sensory areas.
• This high-dimensional, non-categorical coding supports flexible information processing throughout the cortex.
• Findings challenge traditional views of "categorical" neural representations and offer insights into neural computation across the brain.