Neural criticality from effective latent variables

Нейронная критичность, обусловленная эффективными латентными переменными
Ilya M. Nemenman, Audrey Sederberg, Mia C. Morrell
2023-09-12

activity coarse-grainingavalanche criticalitylatent dynamical variablesneural population activitypower-law behavior
Observations of power laws in neural activity data have raised the intriguing notion that brains may operate in a critical state. One example of this critical state is 'avalanche criticality', which has been observed in various systems, including cultured neurons, zebrafish, rodent cortex, and human EEG. More recently, power laws were also observed in neural populations in the mouse under an activity coarse-graining procedure, and they were explained as a consequence of the neural activity being coupled to multiple latent dynamical variables. An intriguing possibility is that avalanche criticality emerges due to a similar mechanism. Here, we determine the conditions under which latent dynamical variables give rise to avalanche criticality. We find that populations coupled to multiple latent variables produce critical behavior across a broader parameter range than those coupled to a single, quasi-static latent variable, but in both cases, avalanche criticality is observed without fine-tuning of model parameters. We identify two regimes of avalanches, both critical but differing in the amount of information carried about the latent variable. Our results suggest that avalanche criticality arises in neural systems in which activity is effectively modeled as a population driven by a few dynamical variables and these variables can be inferred from the population activity.
1
Avalanche criticality occurs without fine-tuning model parameters in both single- and multiple-latent-variable regimes.
2
Multiple dynamical latent variables can generate avalanche criticality in neural population activity.
3
Populations coupled to multiple latent variables exhibit critical behavior across a broader parameter range than populations coupled to a single quasi-static latent variable.
4
The findings suggest avalanche criticality emerges when neural activity is driven by a few dynamical variables that can be inferred from population activity.
5
The study identifies two critical avalanche regimes that differ in how much information they carry about the latent variables.

neural populations coupled to multiple latent dynamical variables

the conditions and mechanisms by which latent dynamical variables generate avalanche criticality, including the resulting information carried about the latent variables

Publication Details
Publication Date
2023-09-12
Journal
Publisher
ISSN
Access Type
Author Information
Authors
Ilya M. Nemenman
Audrey Sederberg
Mia C. Morrell
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%