A Network Architecture for Bidirectional Neurovascular Coupling in Rat Whisker Barrel Cortex

Abstract

10The neurovascular coupling is mostly considered as a master-slave relationship between 11 the neurons and the cerebral vessels: the neurons demand energy which the vessels supply in 12 the form of glucose and oxygen. In the recent past both theoretical and experimental studies 13 have suggested that the neurovascular coupling is a bidirectional system, a loop that includes a 14 feedback signal from the vessels influencing neural firing and plasticity. An integrated model 15 of bidirectionally connected neural network and vascular network is hence required to 16 understand the relationship between the informational and metabolic aspects of neural 17 dynamics . In this study, we present a computational model of the bidirectional neurovascular 18 system in the whisker barrel cortex and study the effect of such coupling on neural activity and 19 plasticity as manifest in the map formation. In this model, a biologically plausible self-20 organizing network model of rate coded, dynamic neurons is nourished by a network of vessels 21 modeled using the biophysical properties of blood vessels. The neural layer which is designed 22 to simulate the whisker barrel cortex of rat transmits the vasodilatory signals to the vessels. 23The feedback from the vessels is in the form of available oxygen for oxidative metabolism 24 whose end result is the ATP necessary to fuel neural firing. The model captures the effect of 25 the feedback from the vascular network on the neuronal map formation in the whisker barrel 26 model under normal and pathological (Hypoxia and Hypoxia-Ischemia) conditions.27 28 29 30 Author Summary 31 32Although neurovascular coupling is typically depicted as a unidirectional influence 33 originating from the neurons and acting on the cerebral vessels, in reality it forms a 34 bidirectional system, consisting of neuronal energy demand signals transmitted to the vessels, 35 and a feedback of metabolic substrates from the vessels, that influence the neural firing and 36 plasticity. We present a computational model of the bidirectional neurovascular coupling in the 37 whisker barrel cortex of rats and study the effect of such coupling on neural activity and 38 plasticity as manifest in the map formation. The model consists of a biologically plausible self-39 organizing dynamic, neural network model and a biophysical vascular network model that 40 nourishes the neural network in the form of oxygen necessary for neural oxidative metabolism. 41The model reproduces the spatio-temporal hemodynamic responses observed in rat whisker 42 barrel cortex. It also demonstrates the essentiality of the vascular feedback on the whisker 43 barrel map formation under normal and pathological conditions. al(9), desynchronized vascular rhythms were found to improve the efficiency of 106 learning in a neural network trained as an autoencoder. Among the aforementioned set of 107 models, there are either detailed biophysical models at the single unit level or abstract models 108 at a network level. There is a need to create detail...

A Network Architecture for Bidirectional Neurovascular Coupling in Rat Whisker Barrel Cortex

“…We would like to thank Ryan T. Philips and Karishma Chhabria for the valuable discussions. This manuscript was released as a Pre-Print at Kumar et al (2019) .…”

Cognition is entangled with metabolism: relevance for resting-state EEG-fMRI

“…Beyond oxygen status, food scarcity (Padamsey et al, 2022), change in neuronal fuel from glucose to ketone bodies (Ma et al, 2007) or circulating immune factors (Tonelli et al, 2005) can also impact neuronal information processing in animal models, highlighting the diversity of metabolic processes that may play a role in normal neural signaling (Watts et al, 2018). These experimental findings are further supported by computational efforts that suggest a role for bidirectional neurovascular coupling in the plasticity of neuronal selectivity (Kumar et al, 2019(Kumar et al, , 2021. Taken together, computational, cellular and systems level findings support the hypothesis that metabolic activity may serve as a generic mechanism to alter the context of information processing dynamics.…”

Artificial Neurovascular Network (ANVN) to Study the Accuracy Vs. Efficiency trade-off in an Energy Dependent Neural Network

“…There have been efforts to construct simplified neuro-energetic models in recent times at the single neuron level. Some of these efforts assume that the effect of energy supply to a neuron can be expressed in abstract terms as regulating the neuron’s threshold of activation: higher (lower) energy leads to a smaller (higher) threshold 45,46 .…”