Diego Candia-Rivera, Luca Faes, Fabrizio De Vico Fallani, Mario Chavez
6 min
The exploration of brain-heart interactions has evolved into a vital milestone for biomarker development, affective computing, and clinical neuroscience. Communication between the brain and heart is essential for interoception, neural homeostasis, and the overall regulation of bodily states. This bidirectional communication operates through multiple pathways spanning genetic factors, circulating extracellular vesicles carrying microRNAs, mechanical pressure waves from heartbeats that influence cerebral blood flow via mechanosensitive ion channels, electromagnetic fields, and classical autonomic pathways mediated by the sympathetic and parasympathetic nervous systems.
A prominent methodological approach involves examining how behavior and neural activity vary across the cardiac cycle, which consists of systole (muscle contraction) and diastole (relaxation). Research demonstrates that perceptual awareness, sensory detection, and motor excitability are modulated by the cardiac phase—for instance, visual and auditory stimuli are more readily detected during diastole, whereas motor responses and active sampling are enhanced during systole. To capture cortical processing of these cardiac inputs, researchers utilize heartbeat-evoked potentials, which are brain potentials time-locked to the R- or T-peaks of the electrocardiogram. While these responses offer windows into cortical monitoring of visceral activity, their computation lacks universal standardization, often facing challenges from cardiac-field artifacts and inter-beat heart rate variability.
Neuroimaging and intracranial electrophysiological recordings consistently highlight a distributed set of brain structures involved in autonomic regulation, collectively known as the central autonomic network. This network includes subcortical nuclei (such as the thalamus, amygdala, and hippocampus), brainstem regions (like the nucleus tractus solitarius), and cortical hubs including the anterior and posterior insular cortices and cingulate cortices. Notably, specific functional trends emerge: sympathetic activations tend to couple with regions of the executive and salience networks (such as the anterior insula and anterior cingulate cortex), whereas parasympathetic activations more frequently associate with regions belonging to the default mode network (such as the posterior cingulate cortex and precuneus).
Quantifying the coupling between brain and cardiac time series requires robust statistical and information-theoretic frameworks. Traditional methods range from linear correlations and spectral coherence to non-linear measures like synchronization likelihood and joint symbolic dynamics. More advanced approaches utilize information dynamics—such as mutual information and the maximal information coefficient—to capture complex, non-linear information exchange between neural oscillations and heart rate variability without requiring arbitrary symbolic transformations. These advanced modeling techniques improve disease stratification and deepen our understanding of how peripheral physiological inputs shape central brain functioning.
The exploration of brain-heart interactions within various paradigms, including affective computing, human-computer interfaces, and sensorimotor evaluation, stands as a significant milestone in biomarker development and neuroscientific research. A range of techniques, spanning from molecular to behavioral approaches, has been proposed to measure these interactions. Different frameworks use signal processing techniques, from the estimation of brain responses to individual heartbeats to higher-order dynamics linking cardiac inputs to changes in brain organization. This review provides an overview to the most notable signal processing strategies currently used for measuring and modeling brain-heart interactions. It discusses their usability and highlights the main challenges that need to be addressed for future methodological developments. Current methodologies have deepened our understanding of the impact of neural disruptions on brain-heart interactions, solidifying it as a biomarker for evaluation of the physiological state of the nervous system and holding immense potential for disease stratification. The vast outlook of these methods becomes apparent specially in neurological and psychiatric disorders. As we tackle new methodological challenges, gaining a more profound understanding of how these interactions operate, we anticipate further insights into the role of peripheral neurons and the environmental input from the rest of the body in shaping brain functioning.
Sam: So how do researchers figure out which direction the influence is flowing? Is the brain driving the heart, or the heart driving the brain?
Alex: Think of it like two musicians playing together — you want to know whether the pianist is following the singer, or the other way around. Researchers use statistical tools that check whether knowing the recent history of one signal helps you predict where the other signal is going next. If past brain activity consistently helps forecast heart activity — better than heart activity alone does — that's evidence the brain is leading. This approach is called Granger causality. Using it, researchers have found that the direction of brain-to-heart signaling can shift between the left and right sides of the brain depending on emotional state.
Sam: What about relationships that aren't straightforward and linear?
Alex: For those, researchers use a different tool called transfer entropy. Instead of assuming a neat, proportional relationship, it uses probability to measure how much information flows from one signal to another, regardless of the shape of that relationship. Applied to sleep, for example, it shows that certain brain rhythms carry substantial two-way information — but that flow weakens as people transition into deep sleep.
Sam: Beyond just pairs of signals, how do scientists model the whole loop at once?
Alex: They build computational simulations where heartbeats and brain oscillations are generated together, each updating based on its own recent history and the current state of the other signal. It lets researchers test hypotheses about the loop as a whole system, rather than just measuring one direction at a time.
Sam: So how does all of this translate into helping actual patients?
Alex: That's where the paper gets into some genuinely meaningful territory. Take depression. It's not purely a brain condition — it frequently overlaps with cardiovascular problems. People with mood disorders often show altered control over the natural fluctuations in their heart rate. And treatments like magnetic brain stimulation, which targets specific brain regions, also produce measurable secondary changes in cardiac activity.
Sam: So monitoring the heart's response could help guide which brain region to target?
Alex: That's the direction being explored, yes — using cardiac signals as a kind of readout to improve treatment precision.
Sam: What about patients with severe brain injuries — people who can't communicate or respond?
Alex: This is one of the more significant potential applications. Doctors often struggle to assess consciousness in patients with severe motor impairments after something like cardiac arrest. But the brain's response to its own heartbeats — how strongly and consistently it registers each pulse — appears to scale with the severity of the injury. Studies cited in the paper suggest those responses can even predict neurological outcomes three months later.
Sam: So the heart becomes a kind of indirect window into the brain's state.
Alex: That's a reasonable way to put it. It offers a meaningful step forward for critical care — though the authors are careful not to overstate it. Many of the signal processing techniques involved are still sensitive to noise and require large amounts of clean data to work reliably. That's a real barrier to routine clinical use.
Sam: So the science is ahead of the clinical infrastructure, in a sense.
Alex: Precisely. The conceptual framework is solid, and the early clinical findings are encouraging — but the authors are clear that substantial work remains before these methods become standard practice. The brain and heart have always been in conversation. We're only beginning to learn how to listen in.
Sam: Thanks, Alex. That's a useful place to land — promising, but honest.
Alex: Thanks for listening to ResearchPod.