Neural Wave Interference in Inhibition-Stabilized Networks

Savel’ev, Sergey; Gepshtein, Sergei

doi:10.3390/ecea-1-c002

Cited by 4 publications

(5 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The generic form of the system is where r E and r I represent the firing rates of the excitatory and inhibitory cells, τ E represents the relaxation time of excitation (in units of the relaxation time of inhibition), l is the location index in the chain, g E , g I are sigmoid functions. The terms W E and W I represent sources of cell activation (from stimulation and from other cells in the network) at location l : where w and represent the weights of connections respectively within and between the motifs (Savel’ev and Gepshtein, 2014).…”

Section: Methodsmentioning

confidence: 99%

“…Equation 3 determines a spatial oscillation (Figure 1C). The intrinsic spatial frequency k n , the rate of spatial decay λ , and parameters that determine the amplitude of oscillation (Γ E , Γ I , Δ E , Δ I ) are functions of the weights of neuronal connections W EE , W EI , W IE , W II and D EE , D EI , D IE , D II (Savel’ev and Gepshtein, 2014).…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Spatially distributed computation in cortical circuits

Gepshtein

Pawar

Kwon

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

The traditional view of neural computation in the cerebral cortex holds that sensory neurons are specialized, i.e., selective for certain dimensions of sensory stimuli. This view was challenged by evidence of contextual interactions between stimulus dimensions in which a neuron's response to one dimension strongly depends on other dimensions. Here we use methods of mathematical modeling, psychophysics, and electrophysiology to address shortcomings of the traditional view. Using a model of a generic cortical circuit, we begin with the simple demonstration that cortical responses are always distributed among neurons, forming characteristic waveforms, which we call neural waves. When stimulated by patterned stimuli, circuit responses arise by interference of neural waves. Resulting patterns of interference depend on interaction between stimulus dimensions. Comparison of these modeled responses with responses of biological vision makes it clear that the framework of neural wave interference provides a useful alternative to the standard concept of neural computation.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Spatially distributed computation in cortical circuits

Gepshtein

Pawar

Kwon

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…where r E and r I represent the firing rates of the excitatory and inhibitory cells,  E represents the relaxation time of excitation (in units of the relaxation time of inhibition), l is the location index in the chain, and g E and g I are sigmoid functions. where w and ~ w represent the weights of connections within and between the motifs, respectively (80). When modulations of neural activity occur on a spatial scale significantly larger than the distance between the nearest network motifs, and at stimulus contrasts for which the system response is far below saturation, the excitatory/inhibitory network can be modeled by a system of nonlinear partial differential equations (derived in the Supplementary Materials) of the form…”

Section: Definition Of the Distributed E/i Networkmentioning

confidence: 99%

Spatially distributed computation in cortical circuits

et al. 2022

Self Cite

View full text Add to dashboard Cite

The traditional view of neural computation in the cerebral cortex holds that sensory neurons are specialized, i.e., selective for certain dimensions of sensory stimuli. This view was challenged by evidence of contextual interactions between stimulus dimensions in which a neuron’s response to one dimension strongly depends on other dimensions. Here, we use methods of mathematical modeling, psychophysics, and electrophysiology to address shortcomings of the traditional view. Using a model of a generic cortical circuit, we begin with the simple demonstration that cortical responses are always distributed among neurons, forming characteristic waveforms, which we call neural waves. When stimulated by patterned stimuli, circuit responses arise by interference of neural waves. Results of this process depend on interaction between stimulus dimensions. Comparison of modeled responses with responses of biological vision makes it clear that the framework of neural wave interference provides a useful alternative to the standard concept of neural computation.

show abstract

“…In our spatially distributed model, the canonical circuit serves as a network motif: a repetitive "node" in a larger network (Figure 1B). This network architecture is a conservative generalization of the canonical circuit: from the spatially local system of Wilson and Cowan to a spatially distributed system suitable for modeling neural response to stimuli distributed both spatially and temporally (Savel'ev and Gepshtein, 2014). That is, the excitatory and inhibitory units of each node are each connected to both the excitatory and inhibitory units of the neighboring nodes, which is a form of connectivity least removed from the canonical model.…”

Section: Introductionmentioning

confidence: 99%

Neural wave interference and intrinsic tuning in distributed excitatory-inhibitory networks

Gepshtein,

Pawar,

Saveliev

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

We developed a model of cortical computation that implements key features of cortical circuitry and is capable of describing propagation of neural signals between cortical locations in response to spatially distributed stimuli. The model is based on the canonical neural circuit that consists of excitatory and inhibitory cells interacting through reciprocal connections, with recurrent feedback. The canonical circuit is used as a node in a distributed network with nearest neighbor coupling between the nodes. We find that this system is characterized by intrinsic preference for spatial frequency. The value of preferred frequency depends on the relative weights of excitatory and inhibitory connections between cells. This balance of excitation and inhibition changes as stimulus contrast increases, which is why intrinsic spatial frequency is predicted to change with contrast in a manner determined by stimulus temporal frequency. The dynamics of network preference is consistent with properties of the cortical area MT in alert macaque monkeys. ContentsPoint-source waves 4 Distributed periodic stimulation 6Distributed spatiotemporal stimulation 8Variation of spatial frequency 9Variation of temporal frequency 11Discussion 12List of Figures 1 The canonical circuit 3 2 Response of the distributed circuit to a point stimulus 5 3 Response of the distributed circuit to increasing contrast 7 4 Intrinsic spatial frequency depends on stimulus temporal frequency 10 5 Intrinsic temporal frequency depends on stimulus spatial frequency 12

show abstract

Neural Wave Interference in Inhibition-Stabilized Networks

Cited by 4 publications

References 22 publications

Spatially distributed computation in cortical circuits

Spatially distributed computation in cortical circuits

Spatially distributed computation in cortical circuits

Neural wave interference and intrinsic tuning in distributed excitatory-inhibitory networks

Contact Info

Product

Resources

About