Measuring direction in the coupling of biological oscillators: A case study for electroreceptors of paddlefish

Abstract

Recently developed methods for estimating directionality in the coupling between oscillators were tested on experimental time series data from electroreceptors of paddlefish, because each electroreceptor contains two distinct types of noisy oscillators. One type of oscillator is in the sensory epithelia, and another type is in the terminals of afferent neurons. Based on morphological organization and our previous work, we expected unidirectional coupling, whereby epithelial oscillations synaptically influence the spiking oscillators of afferent neurons. Using directionality analysis we confirmed unidirectional coupling of oscillators embedded in electroreceptors. We studied the performance of directionality algorithms for decreasing length of data. Also, we experimentally varied the strength of oscillator coupling, to test the effect of coupling strength on directionality algorithms.

Data-driven approach to the estimation of connectivity and time delays in the coupling of interacting neuronal subsystems

“…The most popular approach is based on the concept of Granger causality originally proposed for the analysis of econometric models (Granger, 1969;Geweke, 1982Geweke, , 1984. At present, the concept of Granger causality is widely used in neurophysiology for the detection of directed influences between different activities (Pereda et al, 2005;Brea et al, 2006;Wang et al, 2007;Hlavackova-Schindler et al, 2007;Fine et al, 2010). For instance, the connectivity between different parts of the sensorimotor cortex (Brovelli et al, 2004) or cortico-thalamic interdependencies (Sitnikova et al, 2008) can be detected via Granger causality.…”

Method of estimation of synchronization strength between low-frequency oscillations in heart rate variability and photoplethysmographic waveform variability

“…For example, the approach based on mixing of instantaneous phases by 2π intervals [93] is widely used or the method based on random choice of experimental realizations from single group of patients (i.e. the first signal from one patient and the second from another) [94].…”

Untitled

“…Other methods of surrogate data generation also can be used to control the statistical significance of the results. For example, the approach based on mixing of instantaneous phases by 2π intervals [93] is widely used or the method based on random choice of experimental realizations from single group of patients (i.e. the first signal from one patient and the second from another) [94].…”