Your brain does not process the world in a steady, continuous stream. Every time your heart contracts, roughly once per second, it sends a wave of afferent signals up through the vagus nerve and brainstem to the cortex, competing for the same limited neural resources that process what you see, hear, and feel. A new study published in NeuroImage by Qiaoyue Ren, Simone Schütz-Bosbach, and colleagues at LMU Munich provides some of the clearest evidence yet that this competition is measurable, systematic, and largely invisible to current neuroscience methods.
The finding challenges a fundamental assumption in cognitive neuroscience: that the brain’s responses to external stimuli can be measured independently of its internal physiological state.
The experiment: tagging vision with frequency
Ren and colleagues used a technique called steady-state visual evoked potentials (SSVEPs), flickering visual stimuli at specific frequencies that produce measurable neural oscillations in the visual cortex. Thirty-two participants viewed two groups of moving dots, one flickering at 7.5 Hz and the other at 10 Hz. The dots were superimposed, but the distinct frequencies allowed the researchers to track neural responses to each group independently.
The critical manipulation was timing. One group of dots changed direction during cardiac systole, 290 milliseconds after the ECG R-peak, when the ventricles contract and baroreceptors in the aortic arch and carotid sinuses fire most strongly. The other group changed direction between heartbeats (diastole), when cardiac signals are weakest. A control condition had the direction changes occur at random times, uncoupled from the cardiac cycle. The participants’ task, detecting a brief color change, was unrelated to the heartbeat timing.
The results were striking. SSVEP amplitude was significantly lower for stimuli that coincided with systole and significantly higher for stimuli coinciding with diastole, compared with the uncoupled control. The heartbeat was competing with the visual input, and when the heart signal was strongest, vision lost.
Longer-term effects reinforced the pattern. When visual stimuli were repeatedly coupled with systole across trials, participants showed a larger heartbeat-evoked potential (HEP), the EEG signature of the brain tracking its own heart, and a correspondingly smaller N2 component (a visual-evoked response to the color change). The HEP increase directly predicted the N2 decrease: more interoceptive processing meant less exteroceptive processing.
The baroreceptor hypothesis
The result fits a framework dating back to the 1970s, when John and Beatrice Lacey proposed the baroreceptor hypothesis. Baroreceptors, stretch-sensitive neurons in the aortic arch and carotid sinuses, respond to the blood pressure surge during cardiac systole. Their ascending signals travel through the nucleus tractus solitarius (NTS) in the brainstem, then to the thalamus and onward to the insula and fronto-cortical regions. This pathway broadly suppresses cortical sensorimotor activity during systole, likely as part of a homeostatic mechanism prioritizing internal state monitoring over external input.
Ren and colleagues interpret their findings as support for a “trade-off framework”, spontaneous shifts of attention between interoception (sensing the internal body) and exteroception (sensing the external world), mediated by the cardiac cycle. The heartbeat does not merely accompany perception; it modulates it, beat by beat.
The fMRI problem
The implications extend beyond EEG. Functional MRI (fMRI), the workhorse of human cognitive neuroscience, measures the BOLD (blood-oxygen-level-dependent) signal, which is exquisitely sensitive to blood flow. A review by Kandimalla et al. in NeuroImage (2025, DOI: 10.1016/j.neuroimage.2024.121000) documents the problem: cardiac-linked signals are spatially organized and widespread, overlapping with genuine neural networks, particularly the resting-state networks that are the focus of much fMRI research. Standard correction methods, such as RETROICOR and global signal regression, fail to fully account for these signals because the cardiac contributions are structured, not random.
The result is a systematic confound: differences in functional connectivity between groups or conditions could reflect genuine neural differences, differences in heart function, or both. As Lisa Feldman Barrett, a neuroscientist at Northeastern University, is quoted in the accompanying Science AAAS article: “Heart function is never irrelevant to any task.”
A call for methodological reform
The finding adds urgency to a growing movement within neuroscience to treat physiological signals not as noise to be filtered out but as an integral part of the data. A systematic review by Steinfath et al. published in Psychophysiology (2026, DOI: 10.1111/psyp.70297) examined 132 heartbeat-evoked potential studies and found methods varied so widely, and were so rarely fully reported, that results could not be reliably compared or reproduced across labs. Fewer than 33% of studies had sufficient statistical power. The authors published a reporting checklist and a best-practices database to standardize the field.
For fMRI specifically, the takeaway is that researchers should routinely record ECG and respiration during all scans, randomize stimulus timing across the cardiac cycle, and explicitly report how heartbeat-linked signals were handled in analysis.
What it means for perception and consciousness
The heartbeat’s role in shaping perception may extend beyond simple suppression. Work by Catherine Tallon-Baudry and colleagues (Azzalini et al., Trends in Cognitive Sciences, 2019) has argued that neural responses to heartbeats may be involved in generating the sense of self as an embodied agent, the continuous, first-person perspective that characterizes conscious experience. The heartbeat provides a rhythmic reference frame that anchors the brain’s spontaneous dynamics. If the brain’s response to the world depends on where you are in your cardiac cycle, then consciousness itself may be, at least in part, a conversation between heart and brain.
Clinically, the findings are relevant to anxiety, depression, and PTSD, conditions where brain regulation of the body goes awry, and where interoceptive processing is known to be altered. Sarah Garfinkel at University College London has shown that training people to better sense their cardiac signals can improve emotional regulation in these conditions. Understanding the precise neural mechanisms of the heartbeat-perception trade-off could inform targeted interventions.
Sources:
Ren Q, Marshall AC, Liu J, Schütz-Bosbach S. “Listen to your heart: Trade-off between cardiac interoceptive processing and visual exteroceptive processing.” NeuroImage, Vol. 299, 120808 (2024). DOI: 10.1016/j.neuroimage.2024.120808
Steinfath M, et al. “Heartbeat-Evoked Responses in M/EEG: A Systematic Review of Methods with Suggestions for Analysis and Reporting.” Psychophysiology, Vol. 63, No. 4, e70297 (2026). DOI: 10.1111/psyp.70297
Kandimalla A, et al. “Cardiorespiratory dynamics in the brain: Review on the significance of cardiovascular and respiratory correlates in functional MRI signal.” NeuroImage, Vol. 306, 121000 (2025). DOI: 10.1016/j.neuroimage.2024.121000
[Science AAAS] Williams C. “Your heartbeat quietly shapes how your brain processes information.” Science, June 2026. https://www.science.org/content/article/your-heartbeat-quietly-shapes-how-your-brain-processes-information