Local Neural Wiring May Shape the Brain’s Range of Activity Patterns

The local wiring of neural circuits may help determine how many distinct activity patterns the brain can produce, according to a study published in Nature Neuroscience on September 4, 2026. The research focused on mouse visual cortex and found that recurrent connections within cortical networks were strongly associated with the dimensionality of neural activity.

Dimensionality describes the number of independent activity patterns, or degrees of freedom, expressed by a population of neurons. A higher-dimensional population can occupy a broader range of activity states, while a lower-dimensional population is limited to fewer possible patterns. Because thoughts, perceptions and behaviors arise from groups of neurons rather than isolated cells, researchers view this measure as one way to characterize collective brain dynamics.

David Dahmen, Stefano Recanatesi and colleagues analyzed high-density recordings from the mouse visual cortex collected with Neuropixels probes, which can monitor the electrical activity of hundreds of neurons at the same time. They estimated how many independent patterns were needed to describe the recorded activity and examined whether that number remained stable.

It did not. The team observed that cortical activity moved between states with different dimensionalities over time. The findings suggest that the brain’s activity is not confined to one fixed level of complexity, but can shift among network states that support different numbers of distinct patterns.

The researchers linked this regulation to recurrent synaptic networks—circuits in which neurons are connected through pathways that feed activity back through the network. They further examined physiological measurements of synaptic connections involving more than 32,000 pairs of neurons from mouse and human cortical tissue. This analysis indicated that network motifs associated with dimensionality were present in both species.

That result concerns the prevalence of circuit features in tissue from both species; the activity recordings and dimensionality analysis described in the report were conducted in mouse visual cortex. The study therefore does not establish that the same wiring patterns directly determine the range of activity states in the living human brain.

The authors argue that local circuitry can scale up to influence the degrees of freedom available to larger brain networks. Their methods could be used in future work examining how circuit organization constrains or expands neural activity, including research into activity differences associated with specific conditions and the development of brain-inspired computational models.