Moment-to-Moment Heart–Brain Interactions: How Cardiac Signals Influence Cortical Processing and Time Estimation

Our everyday behavior is affected by sensations originating from our internal organs. These interoceptive sensations are conveyed to the brain, providing moment-to-moment information of the body's physiological state. Interoceptive information is integrated in the brain and influences its function, whether it is consciously perceived or not (Craig, 2009).

Cardiac activity is a key interoceptive source that continuously shapes brain dynamics (Skora et al., 2022). The phases of the cardiac cycle, driven by contraction (i.e., systole) and relaxation (i.e., diastole) of the heart, are encoded by baroreceptors that sense pressure changes in blood vessels. This information travels to the brainstem and from there to various cortical regions, where it is integrated with exteroceptive and other interoceptive information. This enables cardiac activity to influence various cognitive, perceptual, and motor processes and optimize behavior (Marshall et al., 2019; Skora et al., 2022; Fouragnan et al., 2024; Mussini et al., 2024; Paci et al., 2024).

The cardiac activity and the ability to perceive it, often termed interoceptive awareness, are linked to time perception. Indeed, individuals with high interoceptive awareness more accurately estimate elapsed time than those with low interoceptive awareness. This increased accuracy stems from a reduced tendency to contract the duration of (i.e., under-reproduce) intervals lasting several seconds (Meissner and Wittmann, 2011; Richter and Ibáñez, 2021). In addition, the phase of the cardiac cycle influences time perception at the subsecond level, with stimuli presented during diastole being perceived as longer than those presented during systole (Arslanova et al., 2023). Finally, the interbeat interval (IBI), which is a proxy of the vagal tone, increases over time during the encoding of time intervals, and this increase is positively associated with time estimation accuracy (Meissner and Wittmann, 2011). Thus, several cardiac signals influence time estimation at both suprasecond and subsecond timescales.

Electrophysiological evidence provides further support for a link between cardiac interoceptive signals and time perception. Neural processing of cardiac signals is commonly measured by the heartbeat-evoked potential (HEP), an electroencephalographic (EEG) signal time-locked to the R peak of the heartbeat, i.e., the highest positive wave in the electrocardiogram recordings. When participants monitor their own heartbeats, greater HEP amplitudes are associated with dilated time estimation on a suprasecond scale (Richter and Ibáñez, 2021).

The observed link between HEP magnitude and time estimation has been attributed to the “recording” of pulses in an internal accumulator that tracks cardiac activity: more pulses lead to a greater HEP amplitude, which, in turn, leads to subjective dilations of time (Richter and Ibáñez, 2021). This hypothesis, known as the pacemaker-accumulator model (Meissner and Wittmann, 2011), is consistent with the notion of “embodied” time (Craig, 2009) and may explain why perceived duration expands or contracts based on physiological changes, ultimately providing a neural basis for the subjective perception of time. Notably, the neural networks underpinning time estimation and cardiac interoception may share key brain structures. A recent meta-analysis confirms this by demonstrating overlapping activity in the insular cortex, which serves both as a primary neural generator of HEP and as an “internal clock” that dynamically tracks time based on visceral states (Richter and Ibáñez, 2021).

In a recent study, Khoshnoud et al. (2024) explored how cardiac activity influences the encoding and reproduction of time intervals. The authors first aimed to determine whether interoceptive ability influences time estimation performance. They assessed subjective interoceptive awareness with a self-awareness questionnaire and used a heartbeat counting task as an objective measure of interoceptive accuracy. In addition, participants performed a time-interval reproduction task, in which they listened to an auditory stimulus lasting 4, 8, or 12 s and then attempted to reproduce its duration by pressing a button.

In line with previous evidence (Richter and Ibáñez, 2021), participants with higher scores of interoceptive awareness showed greater accuracy in the time-interval reproduction task, especially for the longest interval (12 s), suggesting that internal monitoring supports time estimation. In contrast, no clear association was found between accuracy on the heartbeat counting task and the time-interval reproduction task. The authors noted, however, that previous studies have suggested that the heartbeat counting task is not a reliable test of cardiac interoception (Richter and Ibáñez, 2021).

Furthermore, Khoshnoud et al. (2024) examined the relationship between the HEP amplitude and the accuracy of temporal performance. EEG activity was recorded in participants as they performed the time-interval reproduction task and during a control task, in which they had to press a button a soon as an auditory stimulus ended. To compare time estimation and cortical interoceptive activity, the authors recorded HEP modulation during the encoding and reproduction phases of each interval. During the encoding phase (i.e., listening to the tone), the mean HEP amplitude was smaller for the shortest interval (4 s) than for the two longer intervals; no such effect was observed during the reproduction phase or in the control task. This supports the hypothesis that the HEP reflects a neural accumulator for temporal perception, in line with the pacemaker theory (Richter and Ibáñez, 2021). The authors also assessed the dynamics of the early and late HEP windows during the encoding and reproduction of time. To this aim, HEP were divided into two subcomponents, HEP1 (130–270 ms after the R peak) and HEP2 (470–520 ms). Only during the encoding phase of the 4 s interval, the HEP1 exhibited a linear decrease with the passage of time. In contrast, HEP2 showed a second-by-second monotonic increase during encoding for every interval. During the reproduction phase, only HEP2 was modulated, showing a cumulative increase in magnitude as time passed for all intervals. Interestingly, the magnitude of HEP2 growth negatively correlated with the length of reproduced durations (i.e., under-reproduction): the shorter the reproduced tone, the greater the amplitude of the HEP2 increase. The different HEP1 and HEP2 modulations (decrease vs increase in amplitude) indicate that cardiac signals influence brain activity across distinct temporal windows, suggesting that heart–brain interactions may change between time encoding and reproduction.

In addition, the authors examined changes in the contingent negative variation (CNV), a centroparietal EEG component regarded as a neural marker of time-based responses (Richter and Ibáñez, 2021), during the time-interval reproduction task and the reaction-time task. Khoshnoud et al. (2024)found that the CNV amplitude was greater during the encoding phase than during the control task. These results confirm that the CNV can serve as a neural marker of time processing and support its involvement in cognitive preparation, as proposed by other authors (Marshall et al., 2019; Richter and Ibáñez, 2021; Arslanova et al., 2023). Nonetheless, as also suggested by Khoshnoud et al. (2024), changes in CNV amplitude here may reflect an increased attentional demands during time perception, typically unrelated to interoceptive mechanisms and thus less relevant to investigate brain–heart interaction within this domain. Conversely, HEP modulations during time estimation appear critical to explore the functional connection between the brain processing of visceral signals (such as heartbeats) and temporal cognition (Richter and Ibáñez, 2021).

Overall, Khoshnoud et al. (2024) provide evidence that a stronger awareness of bodily signals (i.e., interoception) is linked to a finer temporal perception. Crucially, they provide neural evidence supporting the influence of cardiac interoceptive signals on time perception, as indicated by distinguished HEP amplitude changes observed in the encoding versus the reproduction of intervals. Interestingly, the specific changes in early (HEP1) and late (HEP2) components during time estimation might reflect two distinctive windows of heart–brain interaction aligned with the two phases of the cardiac cycle. Indeed, these cortical subcomponents of HEP overlap with the time course of systole (∼200–500 ms post-peak R) and diastole (∼500–800 ms post-peak R) phases, respectively. Thus, the striking relationship between HEP2 and time perception might be the result of cortical changes associated with the diastolic phase. In this line, cardiac cycle influences various perceptual processes (for a review, see Skora et al., 2022) and facilitates responses to external stimuli affecting adaptive learning and motor control (Marshall et al., 2019; Fouragnan et al., 2024; Mussini et al., 2024; Paci et al., 2024). This suggests that neurobiological processes in the brain vary according to visceral signals, leading to continual changes in the efficiency of perceptual and cognitive processes (Skora et al., 2022). Future studies should clarify the functional impact of cardiac phases on cortical activity, focusing on the proposed correspondence between the cardiac phases and the HEP windows. Specifically, it should be examined whether early and late components of the HEP reflect periodic baroreceptor firing during the cardiac cycle and thus the cortical representation of systole and diastole.

We reason that, in line with the framework proposed by Skora et al. (2022), the results reported by Khoshnoud et al. (2024) confirm the critical influence of cardiac activity in the modulation of cognitive processes, here the elaboration of time. This is consistent with previous findings that showed associations between changes in the cardiac periods (i.e., IBI), the HEP magnitude, and the resolution of time encoding (Meissner and Wittmann, 2011; Richter and Ibáñez, 2021). Hence, time processing fits well with the predictions of the model described by Skora et al. (2022), adding to other cognitive domains that are functionally affected by cardiac signals. However, further research should aim to reconcile the influence of heart activity on supra- and subsecond time perception, understanding the common denominator (e.g., IBI, cardiac phase, etc.) that predicts the ability to perceive time over distinct scales.

To conclude, the study by Khoshnoud et al. (2024) supports the hypothesis that heart signals shape our sense of time, illustrating that bodily rhythms and internal awareness are highly interconnected with cognitive abilities. This adds to evidence that heart–brain communication extensively refines brain functions, including sensory processing (Skora et al., 2022), decision-making (Fouragnan et al., 2024), motor processes (Marshall et al., 2019), and time estimation.