Several brain disorders are characterized by abnormally strong synchronization of neuronal activity. In Parkinson's patients, permanent high-frequency deep brain stimulation is used to suppress symptoms. To specifically counteract synchronized neuronal activity with a substantially reduced amount of stimulation current, theory-based desynchronizing stimulation techniques were developed, e.g., coordinated reset stimulation. Desynchronizing stimulation may shift adaptive networks from attractors with strong synchronization and strong synaptic coupling to attractors with weak synchronization and weak coupling. This is to cause stimulation effects that persist after cessation of stimulation. Corresponding preclinical and clinical studies reported long-lasting desynchronization and related symptom relief. However, desynchronizing stimulation requires parameters to be adapted to characteristics of the synchronized neuronal activity. Furthermore, desynchronization does not guarantee long-lasting change of network activity. We here present a qualitatively different approach to induce long-lasting, sustained changes of neuronal network dynamics: decoupling stimulation. Instead of primarily desynchronizing neuronal activity, decoupling stimulation employs synaptic plasticity mechanisms to specifically decouple neuronal networks. In this way, neuronal networks get robustly shifted to attractors with desynchronized neuronal activity. We present a theoretical framework that explains how neuronal responses to single stimuli as well as to spatiotemporally correlated stimulus sequences impact on network connectivity. This provides a theoretical base for designing effectively decoupling stimulation protocols. To overcome limitations of primarily desynchronizing stimulation, we present a random reset stimulation protocol, which uses spatiotemporal stimulus randomization to effectively decouple neurons. Theoretical predictions of random reset-induced decoupling as opposed to desynchronization-induced decoupling achieved by coordinated reset stimulation are compared to simulations of networks of integrate-and-fire neurons with spike-timing-dependent plasticity. Decoupling and related long-lasting desynchronization effects achieved by random reset stimulation are more robust with respect to parameter changes than those for coordinated reset stimulation. For both random reset and coordinated reset stimulation, simulation results and theoretically predicted decoupling rates show good quantitative agreement for sufficiently strong stimulation amplitudes. Intriguingly, single stimulus-related mechanisms may have a stronger decoupling impact than stimulus sequence-related mechanisms. We discuss scope and limitations of our decoupling approach for different types of synaptic plasticity and its application to deep brain stimulation.

• Received 12 December 2019
• Accepted 5 June 2020

DOI:https://doi.org/10.1103/PhysRevResearch.2.033101

Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article's title, journal citation, and DOI.

Published by the American Physical Society