Common Input Explains Higher-Order Correlations and Entropy in a Simple Model of Neural Population Activity

Macke, Jakob H.; Opper, Manfred; Bethge, Matthias

doi:10.1103/physrevlett.106.208102

Cited by 89 publications

(277 citation statements)

References 16 publications

Supporting

Mentioning

224

Contrasting

Unclassified

Order By: Relevance

“…9B). Consistent with our results shown in Figure 9A and previous findings (Macke et al, 2009(Macke et al, , 2011, this difference was more pronounced when rates were low and pairwise correlations were strong, which is typically the case for nLFPs in ongoing activity .…”

Section: Accurate Approximation Of Instantaneous Lfp Patterns By the supporting

confidence: 82%

“…Theoretical studies proposed nonlinear, e.g., threshold, operations as a potential origin for higher-order interactions (Amari et al, 2003;Macke et al, 2011). Therefore, to investigate whether higher-order interactions in neuronal avalanches can be explained by thresholding, we fit our data with the DG model (Amari et al, 2003;Macke et al, 2009Macke et al, , 2011.…”

Section: Accurate Approximation Of Avalanche Patterns By a Simple Parmentioning

confidence: 99%

“…Therefore, to investigate whether higher-order interactions in neuronal avalanches can be explained by thresholding, we fit our data with the DG model (Amari et al, 2003;Macke et al, 2009Macke et al, , 2011. Like the Ising model, the DG model is a stochastic model without free parameters, i.e., all parameters are determined by the observed nLFP rates at each electrode and pairwise correlations between nLFPs at different sites.…”

Section: Accurate Approximation Of Avalanche Patterns By a Simple Parmentioning

confidence: 99%

“…The dichotomized Gaussian (DG) model (Amari et al, 2003;Macke et al, 2009Macke et al, , 2011) has a threshold operation based on multivariate Gaussian variables: y i ϭ 1 when u i Ͼ 0 and y i ϭ 0 otherwise, where u ϭ (u 1 , u 2 , …, u n ) ϳ N (␥, ⌳) and ␥ is the mean and ⌳ is the covariance of the Gaussian variables, respectively. To match the rate, r, and covariance, ⌺, of the observed binary variables, ␥ and ⌳ need to be adjusted according to ␥ i ϭ ⌽ Ϫ1 (r i ) and ⌳ ij as the solution for…”

Section: Rates and Pairwise Correlations Smaller Than 10mentioning

confidence: 99%

“…Here we show that pairwise interactions alone are not sufficient to identify neuronal avalanche dynamics (Beggs and Plenz, 2003), measured as ongoing cortical activity in awake monkeys based on the LFP. We then demonstrate that the incorporation of a specific structure of higher-order interactions that results from thresholding (Amari et al, 2003;Macke et al, 2011) improves the accuracy of reconstructing neuronal avalanche dynamics by up to two orders of magnitude. The same method is then shown to significantly improve the approximation of ongoing spiking activity in awake monkeys as well as visually evoked spiking responses in anesthetized cats.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Higher-Order Interactions Characterized in Cortical Activity

Yu¹,

Yang²,

Nakahara³

et al. 2011

J. Neurosci.

201

267

View full text Add to dashboard Cite

In the cortex, the interactions among neurons give rise to transient coherent activity patterns that underlie perception, cognition, and action. Recently, it was actively debated whether the most basic interactions, i.e., the pairwise correlations between neurons or groups of neurons, suffice to explain those observed activity patterns. So far, the evidence reported is controversial. Importantly, the overall organization of neuronal interactions and the mechanisms underlying their generation, especially those of high-order interactions, have remained elusive. Here we show that higher-order interactions are required to properly account for cortical dynamics such as ongoing neuronal avalanches in the alert monkey and evoked visual responses in the anesthetized cat. A Gaussian interaction model that utilizes the observed pairwise correlations and event rates and that applies intrinsic thresholding identifies those higher-order interactions correctly, both in cortical local field potentials and spiking activities. This allows for accurate prediction of large neuronal population activities as required, e.g., in brain-machine interface paradigms. Our results demonstrate that higher-order interactions are inherent properties of cortical dynamics and suggest a simple solution to overcome the apparent formidable complexity previously thought to be intrinsic to those interactions.

show abstract

Section: Accurate Approximation Of Instantaneous Lfp Patterns By the supporting

confidence: 82%

Section: Accurate Approximation Of Avalanche Patterns By a Simple Parmentioning

confidence: 99%

Section: Accurate Approximation Of Avalanche Patterns By a Simple Parmentioning

confidence: 99%

Section: Rates and Pairwise Correlations Smaller Than 10mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Higher-Order Interactions Characterized in Cortical Activity

Yu¹,

Yang²,

Nakahara³

et al. 2011

J. Neurosci.

201

267

View full text Add to dashboard Cite

show abstract

Critical Exponents, Universality Class, and Thermodynamic “Temperature” of the Brain

Yang

Shriki

et al. 2014

Criticality in Neural Systems

View full text Add to dashboard Cite

Anterior cingulate cortex differently modulates frontoparietal functional connectivity between resting‐state and working memory tasks

Zhang

Biswal

2020

Human Brain Mapping

View full text Add to dashboard Cite

The brain frontoparietal regions and the functional communications between them are critical in supporting working memory and other executive functions. The functional connectivity between frontoparietal regions are modulated by working memory loads, and are shown to be modulated by a third brain region in resting‐state. However, it is largely unknown whether the third‐region modulations remain the same during working memory tasks or were largely modulated by task demands. In the current study, we collected functional MRI (fMRI) data when the subjects were performing n‐back tasks and in resting‐state. We first used a block‐designed localizer to define the frontoparietal regions that showed higher activations in the 2‐back than the 1‐back condition. Next, we performed physiophysiological interaction (PPI) analysis using left and right middle frontal gyrus (MFG) and superior parietal lobule (SPL) regions, respectively, in three continuous‐designed runs of resting‐state, 1‐back, and 2‐back conditions. No regions showed consistent modulatory interactions with the seed pairs in the three conditions. Instead, the anterior cingulate cortex (ACC) showed different modulatory interactions with the right MFG and SPL among the three conditions. While the increased activity of the ACC was associated with decreased functional coupling between the right MFG and SPL in resting‐state, it was associated with increased functional coupling in the 2‐back condition. The observed task modulations support the functional significance of the modulations of the ACC on frontoparietal connectivity.

show abstract

Common Input Explains Higher-Order Correlations and Entropy in a Simple Model of Neural Population Activity

Cited by 89 publications

References 16 publications

Higher-Order Interactions Characterized in Cortical Activity

Higher-Order Interactions Characterized in Cortical Activity

Critical Exponents, Universality Class, and Thermodynamic “Temperature” of the Brain

Anterior cingulate cortex differently modulates frontoparietal functional connectivity between resting‐state and working memory tasks

Contact Info

Product

Resources

About