• biomedical signal processing
• brain-heart interplay
• Physiological modeling

In the course of this doctoral work, methodological advancements were pursued in the field of brain-heart interplay, for the understanding of the role of such interplay in different aspects of human cognition. The role of the body on brain functions and consciousness was hypothesized in ancient philosophical postulates and emphasized in the explanation of human emotions. Nevertheless, the role of these visceral signals has been undervalued, as brain imaging techniques have drastically improved, the evolving neuroscientific research has focused mostly in exploring neural dynamics at brain level only. The scientific evidence has shown a steady interaction between central and autonomous nervous systems, which is disrupted in several pathological conditions. For instance, on the effects of brain trauma on cardiac activity, or on the assessment of cardiac activity as a marker of psychological and psychiatric conditions. More recent evidence depicts that brain-heart interplay predicts or reflects perception at different tasks, supporting hypotheses of an embodied view of cognition. In predictive coding framework, the brain integrates interoceptive and exteroceptive information to perform predictions of the ongoing environmental changes. Other hypotheses state that visceral activity is inherent for the brain to experience subjectivity and it would be required to raise awareness of inner mental and physical states, and the external world.
My work in this doctoral project provides with methodological advancements to assess in a non-invasive manner the ongoing brain-heart interactions, which can be applied to different cognitive paradigms. I investigated the impact of using different electroencephalographic references on the outputs of different markers of brain-heart interplay. I provide with guidelines for a robust estimation, showing that an average reference from a wide scalp coverage is a reliable estimate of the electrical reference. I developed a new Integral Pulse Frequency Modulation Model for synthetic data generation of RR series, namely Sympatho-Vagal Modulation Model. The introduced model leverages on orthonormal Laguerre expansions of RR series, instead of the classical sinusoidal expansion, which addresses the problem of an inaccurate estimation of sympathetic and parasympathetic outflow. I propose a methodology to measure brain-heart interplay, embedding the aforementioned model with a Markovian model of synthetic data generation of EEG series. The mutual modulations between EEG and heart rate oscillations are estimated through these coupled models.
I focus on the brain-heart interplay in two different conditions: under emotion elicitation and cold-pressor test. From our results, I show that ascending cardiac inputs to the brain, as measured with EEG, initiate and shape arousal at cortical level in both emotional and physical arousing conditions. These results are relevant in a methodological and neuroscientific point of view, as the proposed framework may help to understand the brain-heart interplay at different conditions, but also provide with evidence supporting long-lasting hypotheses of physiological feelings, in which the visceral activity prompts the emotional processing.
The present work is structured in five chapters: Introduction, reviewing the theoretical landscape of embodied cognition and the experimental evidence supporting it; Methodological description and pre-processing, in which I detail the methods used for the data analysis, and the state-of-the-art on the analysis of brain-heart interplay is reviewed; Methodological advancements in the measurement of brain-heart interplay, in which the proposed methodology for synthetic data generation of RR series and the new framework to estimate brain-heart interplay is explained; Results and Discussion, where the results on brain-heart interplay during emotion elicitation and cold-pressor test are presented and discussed. These results are presented in a critical point of view, meaning that the proposed analysis has advantages, but also limitations, which can be addressed in future research to continue the understanding of why visceral activity may be necessary for the brain functions we observe in our daily lives.