Hebb's Vision: The Structural Underpinnings of Hebbian Assemblies

Area of Science:

• Neuroscience
• Systems Neuroscience
• Computational Neuroscience

Background:

• Donald Hebb proposed neuronal assemblies as fundamental units of perception and cognition in 1949.
• Identifying the structural basis of these assemblies has been limited by the lack of large-scale structure-activity maps.

Purpose of the Study:

• To investigate the relationship between neural activity patterns and underlying neural network structure.
• To test Hebb's theory of neuronal assemblies using novel integrated physiological and structural data.

Main Methods:

• Analysis of a novel dataset linking in vivo optical physiology with postmortem electron microscopy (EM) connectivity.
• Extraction of neural assemblies from fluorescence traces using higher-order correlations in neural activity.
• Comparison of assembly properties with size-matched random neuronal ensembles.

Main Results:

• Neural assemblies showed more reliable responses to natural stimuli and superior visual stimulus decoding compared to random ensembles.
• Neurons within assemblies were significantly more integrated into the overall structural network.
• Contrary to predictions, direct excitatory connections within assemblies were not strengthened; instead, indirect feed-forward inhibition between assemblies was significantly stronger.

Conclusions:

• Neural assemblies are functional components of perception and cognition.
• The structural organization of assemblies is characterized by mutual inhibition between different assemblies, challenging previous assumptions.