Dissecting the integrated information of cardiovascular and cardiorespiratory systems at rest and during...

Objective.In the framework of integrated information theory, different measures have been proposed to quantify the information integrated in a dynamic system composed by interconnected units, each capturing a different aspect of the system connections. While integrated information measures have been traditionally applied to brain dynamics, in this study we propose their implementation for the investigation of the short-term regulation of cardiovascular (CV) and cardiorespiratory (CR) systems reflected by the spontaneous variability of heart period (HP), systolic arterial pressure (SAP) and respiration (RESP).Approach.We selected two integrated information measures related to well-known dynamic properties of CV and CR systems: whole-minus-sum (WMS) quantifying the information stored in the whole system above and beyond the information stored in its units, and causal density (CD) that handles integration in terms of the information transferred between the system units. WMS and CD measures were computed from the parameters of vector autoregressive models fitted respectively on the bivariate time series of{HP, SAP}and{HP, RESP}, obtained from healthy young subjects at rest, during head-up tilt and during a mental arithmetic task.Main results.We found that CV and CR systems dynamically integrate significant amounts of information, which undergo modulations in response to the tasks performed, especially during tilt, proving that orthostatic stress significantly reduces the degree of cooperation between the parts of the systems. A different response to stress was observed for CD and WMS measures, related to the changes in the information storage and transfer induced by the orthostatic and mental challenges and their physiological interpretation.Significance.The possibility offered by the integrated information measures, in combination with their constituent contributions, allowed us to disentangle in a comprehensive way the statistical relationships underlying the CV and CR dynamics, and to investigate for the first time the concept of integration in the CV and CR systems.