Neural and Cardiac Contributions to Perceptual Suppression During Cycling

Neural and Cardiac Contributions to Perceptual Suppression During Cycling | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL Psychophysiology This is a preprint and has not been peer reviewed. Data may be preliminary. 27 January 2025 V1 Latest version Share on Neural and Cardiac Contributions to Perceptual Suppression During Cycling Authors : Aishwarya Bhonsle 0000-0002-7848-3845 and Melanie Wilke [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.173796146.68913752/v1 Published Psychophysiology Version of record Peer review timeline 574 views 195 downloads Contents Abstract Introduction Methods Experimental procedure EEG acquisition and preprocessing ECG acquisition, preprocessing, and instantaneous heart rate analysis Behavioural analysis Alpha amplitude analysis Heartbeat-evoked potentials analysis Statistical analyses Results Alpha amplitude analysis Heart rate analysis Heartbeat-evoked potentials analysis Discussion References Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Exercise influences visual processing and is accompanied by neural and physiological changes in the body. Yet, the underlying mechanisms by which neural and physiological responses to exercise impact ensuing perception remain poorly understood. Particularly, the effects of exercise-induced cardiac changes on visual perception and electrophysiological activity are unclear. Here, we aimed to investigate the relationship between conscious visual perception, neural activity, and cardiac responses during exercise. Thirty healthy participants performed a perceptual suppression task while engaging in light-intensity stationary cycling, with EEG and ECG activity recorded simultaneously. Our study shows that the probability of perceptual suppression decreased during cycling. Parieto-occipital alpha amplitudes (8–12 Hz) also decreased during cycling, but this reduction did not correlate with the decrease in perceptual suppression. Additionally, cycling also decreased heartbeat-evoked potential (HEP) amplitudes, indicating altered neural processing of cardiac signals during exercise and a potential influence of cardiac physiology on HEPs. However, these exercise-induced changes in HEP amplitudes did not predict perceptual outcomes. Moreover, changes in heart rate in response to cycling did not correlate with changes in perceptual suppression rates, pre-stimulus alpha or HEP amplitudes. These findings indicate that while exercise modulates conscious visual perception, the associated changes in alpha activity, heart rate, and HEPs do not fully explain this effect. Our results highlight the complex relationship between interoceptive processing and mechanisms underlying the perception of external stimuli during exercise. Introduction A single bout of exercise can produce positive effects on cognitive function 1–4 , alongside various neural 5,6 and bodily physiological responses 7 . Studies of human visual processing during physical activity reveal that exercise-induced neural changes are linked to altered perceptual dynamics, which differ markedly from those observed at rest 8–11 . Among these neural changes, cortical oscillatory activity across the entire frequency spectrum and various brain regions shows consistent modulation during acute bouts of physical activity, with findings in the alpha band (8-12 Hz) being the most consistent, though the direction and location of these effects vary (see refs. 5,6 for reviews). These neural responses are highly context-dependent, as variations in the combinations of perceptual tasks and types of physical activity contribute to differences in the observed outcomes. For example, naturalistic walking has been linked to a decrease in alpha power, which is thought to reflect reduced inhibitory processes that typically suppress peripheral visual input 8 . During an oddball task, moderate-to-high intensity cycling led to a smaller decrease in parieto-occipital alpha power in response to nontarget stimuli compared to light exercise, while detection accuracy remained consistent across intensities 12 . In contrast, increased parieto-occipital alpha has also been observed during cycling, coinciding with reduced accuracy in an orientation discrimination task 9 . Independent of perceptual task demands, broadband decreases in alpha activity have been observed during cycling and walking, with more pronounced reductions during walking 13 . Conversely, increases in alpha activity during cycling have also been observed throughout the brain 14 , though most frequently across anterior regions 6,15–17 . In addition to these neural responses, exercise elicits an increase in cardiac activity 7 . While the effects of exercise on oscillatory neural activity and visual perception are the subject of ongoing investigation, the influence of exercise-induced cardiac changes on neural activity and subsequent perception remains largely understudied. Typically, studies investigating the effects of physical activity used heart rate as a metric of exercise intensity 18 , while EEG analyses consider heart activity as a source of non-cerebral artifacts, especially under physical exertion. However, it has been theorized 5,6 that exercise-induced increases in alpha activity might arise from cortical inhibition driven by brainstem and subcortical activation linked to cardiovascular regulation 19,20 . This proposal brings together two key concepts: the baroreceptor hypothesis 20–24 and alpha activity as an index of cortical excitability 25–29 . The baroreceptor hypothesis suggests that changes in cardiovascular activity influence cortical excitability via inhibitory afferent feedback from baroreceptors, which are mechanoreceptors in the heart and blood vessels that relay information to the brain about the timing and strength of cardiac contractions 20–24 . According to this hypothesis, higher heart rates, which cause the heart to pump more frequently and increase baroreceptor firing, should lead to an increase in inhibitory feedback and a subsequent reduction in cortical excitability. In parallel, alpha activity is regarded as an index of cortical excitability, with higher levels of alpha levels reflecting reduced cortical excitability 25–29 and visual attentiveness 30–32 . This relationship between alpha activity and cortical excitability provides a potential neural mechanism by which exercise-induced cardiovascular changes, as predicted by the baroreceptor hypothesis, could influence brain function, although this has yet to be directly investigated. Changes in heart rate are also thought to be a key physiological response mediating the effect of exercise on perceptual and cognitive performance 2,3 . Although this relationship is not well characterized in the context of exercise, existing evidence highlights a link between heart rate dynamics and visual perception. For instance, individuals with lower heart rates demonstrate increased visual stimulus detection accuracy 33 . Additionally, studies employing perceptual paradigms performed at rest reveal that heart rate changes occur both leading to and after perception of an external stimulus 20–22,34–42 . Specifically, heart rate typically decelerates in anticipation of an upcoming stimulus, a pattern thought to reflect sustained attention and preparatory processes 20,41,42 , and then accelerates again, following stimulus detection or response registration. These instantaneous adjustments in heart rate are thought to fine-tune the balance between internal bodily signals (interoception) and external sensory inputs (exteroception), thereby facilitating perception and action 42 . The cardiac deceleration has been proposed to reduce the inhibitory influence of baroreceptor activity, enhancing attention to external stimuli, improving sensory processing and perception. Consistent with this notion, preliminary evidence shows that cardiac deceleration tracks active attention during binocular rivalry 43 . Conversely, the heart rate acceleration following response registration restores the balance between interoceptive and exteroceptive processing and prepares the system for action. Interestingly, this competition between interoception and exteroception, independent of cardiovascular mechanisms, has also been discussed in the context of exercise, where shifts in attentional focus between these domains are thought to modulate perception of fatigue 44,45 . However, while resting heart rates and instantaneous heart rate changes have been linked to perception of visual stimuli and thought to reflect an interoceptive-exteroceptive trade off, how sustained heart rate increases induced during exercise influence visual perceptual dynamics, particularly in relation to alpha oscillations, remains unexplored. Moving beyond heart rate and alpha activity changes, the heartbeat-evoked potential (HEP) is a neural measure that has not yet been investigated in the context of exercise. The HEP represents transient neural activity observed when electrophysiological data is time-locked to heartbeats, thought to serve as a marker of the cortical processing of cardiac information 46–49 . HEP amplitudes are increased during interoceptive tasks such as heart beat counting 50,51 . HEP amplitude has been shown to predict conscious perception of visual 37 and somatosensory 52 stimuli at threshold, further linking interoceptive and exteroceptive processing. Investigating the effect of cycling on HEPs could offer valuable insights into how exercise modulates heart-brain coupling, potentially illuminating interactions between cardiac and cortical responses. Furthermore, examining whether cycling-related modulation of HEPs influences conscious perception could provide additional insight into the contribution of interoceptive processing to exteroceptive perceptual awareness. In this study, we investigated the neural and cardiac changes through which exercise influences conscious visual perception in a bistable perceptual suppression paradigm. Specifically, we examined how stationary cycling modulates oscillatory brain activity, heart rate, and heart-brain coupling, and how these responses potentially influence conscious perception in a Generalized Flash Suppression (GFS) paradigm. GFS is a perceptual suppression paradigm in which a salient target is rendered subjectively invisible upon presentation of a moving surround 53 . Previous work employing GFS performed at rest has shown that parieto-occipital alpha amplitudes in the second prior to the motion onset were significantly decreased preceding target disappearances, compared to when the target remained visible 54 . Therefore, in the current study, exercise-induced changes in alpha oscillations may serve not only as an index of cortical excitability, but may also predict perceptual suppression. How might perceptual suppression and related pre-stimulus parieto-occipital alpha activity be altered during cycling? Evidence from prior studies suggests that alpha activity often increases during exercise 6,9,14–17 , and extending the baroreceptor hypothesis to the exercise context further links elevated heart rates during exercise to decreased cortical excitability and increased alpha oscillations 5,6,19,20 . However, decreases in alpha during physical activity have also been reported in some studies 8,13 , presenting an alternative possibility. Based on these observations and previous GFS findings 54 , we hypothesized that decreased pre-stimulus, parieto-occipital alpha amplitudes will be associated with higher subjective target suppression rates, whereas increased pre-stimulus, parieto-occipital alpha amplitudes will correspond to higher target visibility rates. We also investigated how cycling modulates HEPs. Given the heightened exteroceptive attention required to perform a perceptual and motor task simultaneously, we hypothesized that interoceptive processing will be reduced, reflected by decreased HEP amplitudes during cycling. Additionally, we explored if HEPs predict perceptual suppression. Given that perceptual suppression is associated with lower pre-stimulus occipital alpha activity, which indicates increased external attention 30–32 , increased exteroceptive attention might be correlated with target suppression during GFS. If increased exteroceptive processing competes with interoceptive processing 42 , we might expect lower HEP amplitudes when the target disappears. Methods Participants Forty-nine healthy volunteers, mostly university students, were initially recruited for the current study. To be eligible for participation, individuals were required to have no history of neurological, psychological, cardiovascular, respiratory, metabolic, endocrine, immune, or substance abuse disorders. Further eligibility criteria included not taking regular medications that might alter cardiovascular, autonomic, or cognitive functioning, not being pregnant, and not concurrently participating in pharmacological or stimulation studies. All subjects were also required to have normal or normal-to-corrected vision. Of the recruited participants, two did not complete all sessions of the experiment. During the practice session on day 1, five participants never perceived the target disappearing, and two perceived it disappearing in every trial; all seven were excluded from further participation in the study on day 2. Ten subjects were excluded from further analysis due to noisy EEG data, largely caused by movement artifacts during the cycling conditions, as their recordings required more than 15% of channels to be interpolated. The final cohort consisted of 30 subjects (14 male, 16 female; mean age \(\pm\) SD: 23.93\(\pm\) 2.41 years, age range: 20 – 29 years). One subject was excluded from the HEP analysis due to a lack of trials that met the inclusion criteria for this analysis in one condition. The study was conducted in accordance with the ethical guidelines of the Declaration of Helsinki. The experimental procedure was approved by the ethics committee of the University Medicine Göttingen (UMG, Germany). All subjects gave written informed consent before the study and were paid for their participation. Stimuli and task The visual stimuli were programmed in MATLAB R2015b (The MathWorks Inc., Natick, MA, USA) using Psychtoolbox-3 55,56 . Each trial began with a white fixation cross on a black background that remained visible throughout the trial ( Figure 1A ). After two seconds of central fixation, the target, a red disk with a diameter of 3° of visual angle, was presented in the left visual hemifield, positioned 7° horizontally and 3° vertically from the centre. After an additional two seconds, a random dot motion (RDM) pattern of moving blue dots (dot diameter = 0.08°; dot speed = 10°/s; dot density = 1.0 dot/deg 2 ) appeared for two second. A 0.5° buffer zone, set to the background colour, was maintained between the target and RDM pattern. The target and RDM pattern were each presented monocularly using red-blue anaglyphic glasses. In a subset of trials, the presentation of the RDM stimulus resulted in the subjective disappearance of the target. Following each trial, a blank screen was shown for three seconds as the inter-trial interval (ITI). Subjects were instructed to maintain fixation on the fixation cross throughout the trial and to use a button box (4 Button Curve Right; Current Designs Inc., Philadelphia, PA, USA) to report their perception of the target. Using the index finger of their dominant hand (27 right-handed, 3 left-handed as assessed by the short version of the Edinburgh Handedness Inventory 57 ), participants pressed the response button upon target onset. They were instructed to hold the button down as long as the target was visible, releasing it if the target disappeared. If the target reappeared before the end of the trial, they were asked to report this by pressing and holding the button again. Two types of control trials ( Figure 1A ) were intermixed with the experimental trials: catch trials and RDM OFF trials. In the catch trials, the target was physically removed upon onset of the RDM pattern, while in the RDM OFF trials, the RDM pattern was never presented. ( A ) Time courses of GFS, catch and RDM OFF trials. GFS trials began with a 2 s central fixation, followed be the onset of a salient red target in the upper left visual hemifield. After 2 s of target presentation, a random dot motion (RDM) stimulus was presented for a further 2 s, resulting in the disappearance of the target in a subset of trials. During catch trials, the target was physically removed at RDM onset. During RDM OFF trials, the RDM pattern was never presented. Subjects reported their detection of the target by pressing and holding the button for as long as they could see it, and releasing the button to indicate its disappearance. There was a blank screen presented for 3 s as the inter-trial interval (ITI). ( B ) The three rest (R) blocks each consisted of 61 trials: 44 GFS, 10 catch, 7 RDM OFF . The four cycling blocks (2 low-resistance (L) and 2 high-resistance (H)) each consisted of 90 trials: 65 GFS, 15 catch, 10 RDM OFF . Each cycling block began with a 3-minute warm-up period before the GFS paradigm was presented. ( C ) The experimental setup consisted of the subject seated at a desk facing a computer screen where the perceptual task was presented. They reported their perceptual responses using a button box placed on the desk. They wore red-blue anaglyphic glasses and their head was positioned on a chin rest, and stabilised by positioning supports on either side and against their forehead (top inset). During the cycling blocks, they cycled using an ergometer placed under the desk (bottom inset). EEG and ECG were recorded simultaneously. The subjects performed GFS under three exercise intensity conditions ( Figure 1B ): at rest (R), and during low-resistance (L) or high-resistance cycling (H). The two cycling conditions were performed using an ergometer (Sportstech DFX100; Sportstech, Berlin, Germany) placed under the desk. According to the American College of Sports Medicine guidelines 18 on exercise intensity, categorised by percentage of maximal heart rate (HRmax, estimated by the formula 211-0.64*age) 58 , the low-resistance condition would be classified as very light (<50% HRmax), while the high-resistance condition would be classified as light (50-60% HRmax) intensity. Each rest block ( Figure 1B ) comprised 61 trials (44 GFS, 10 catch, and 7 RDM OFF ) and lasted approximately 12 minutes. Each cycling block ( Figure 1B ) comprised 90 trials (65 GFS, 15 catch, and 10 RDM OFF ) and lasted a total of 15 minutes, beginning with a 3-minute warm-up period without a perceptual task, followed by 12 minutes of simultaneous cycling and perceptual task performance. During the initial 3-minute warm-up period for each cycling block, participants were instructed to establish a cadence of 60 rotations per minute (rpm), guided by an on-screen clock displaying seconds to help them match the pace. When the clock was replaced by the perceptual task, participants were instructed to maintain this cadence of 60 rpm to the best of their abilities until the end of the block. Experimental procedure The study was conducted over two sessions, spaced no more than 10 days apart. Overall, the first session lasted approximately 1-1.5 hours, and the second session lasted 3.5-4 hours. In session 1, participants completed a demographic survey, a questionnaire to assess interceptive awareness (MAIA) 59 , and the Ishihara colour vision test 60 . For inclusion in the study, subjects were required to have 19 correct identifications out of 20 Ishihara plates. Following this, they received instructions for the task and completed one practice block of each of the three exercise intensity conditions. Participants were instructed to abstain from caffeine, alcohol, or other mind-altering substances for 3-4 hours prior to session 2, and from heavy alcohol consumption for 24 hours before the session. In this experiment, EEG and ECG data were recorded simultaneously. Breathing data was recorded using a respiration belt measuring thoracic/abdominal movements (Respiration Belt MR; Brain Products GmbH, Gilching, Germany) but was not analysed in the context of this paper. Following EEG and ECG preparation, an 8-minute baseline recording with eyes closed was taken, with blood pressure measured at the 7-minute mark. Participants were then reminded of the task instructions and proceeded to perform the experiment. To perform the experimental task ( Figure 1C ), subjects were seated in front of a 60x34 cm computer screen with a resolution of 1920\(\times\) 1080 pixels and 60 Hz refresh rate (BenQ XL2411T; BenQ, Taipei, Taiwan), and the eye to screen distance was 70 cm. They placed their head on a chin rest and it was stabilised by positioning supports on each side of the head and one to lean their forehead against (HeadLock Ultra Precision Head Positioner; Arrington Research, Scottsdale, AZ, USA), to minimise head movement. The lights in the recording room were turned off during the experiment and additional curtains were used to protect the subjects from extraneous light. The experiment consisted of seven GFS blocks, with cycling (low-resistance – L, high-resistance - H) and rest (R) blocks interleaved in one of two sequence orders, either L-R-H-R-L-R-H or H-R-L-R-H-R-L, pseudorandomly assigned for each participant. Subjects were allowed adequate breaks between blocks and special care was taken to allow heart rates to return to baseline after each cycling block. Baseline heart rate was defined as the rate observed during the initial baseline recording and blood pressure measurement. Live heart rate was monitored throughout the breaks using an armband (TikrFit; Wahoo Fitness, Atlanta, GA, USA). After the experimental blocks, another 8-minute baseline recording with eyes closed was taken. EEG acquisition and preprocessing EEG activity was recorded from 64 electrodes distributed over the head according to the international 10-20 system (actiCap Snap, BrainAmp MR, BrainVision Recorder; Brain Products GmbH, Gilching, Germany). Electrode impedances were kept below 20 kΩ throughout the experiment. The data were recorded at a sampling rate of 1000 Hz. EEG data were preprocessed and analysed using the FieldTrip toolbox 61 and custom-written software in MATLAB R2015b (The MathWorks Inc., Natick, MA, USA). The data were down-sampled to 256 Hz and band-pass filtered between 0.5 and 110 Hz. A band-stop filter was applied to remove 50 Hz line noise. The continuous data was then segmented into trials. Trials containing muscle artefacts, jumps or clipping artefacts were identified automatically, visually inspected and then rejected when necessary. Overall, 18% of trials were excluded. An independent component analysis (ICA) was performed to identify eye movement related artefacts and the relevant components were removed. The data were the re-referenced to a common average reference. If there were noisy channels that required interpolation, they were removed before the data were re-referenced. They were then interpolated with the re-referenced data. ECG acquisition, preprocessing, and instantaneous heart rate analysis The ECG montage used in this study was based on the one described by Petzschner et al, 2019 51 . Two ECG signals were acquired using two electrodes placed of the left and right clavicle (active electrodes), two electrodes placed at the left and the right hip/abdominal (reference electrodes), and a ground electrode placed between the shoulder blades (BrainAmp ExG; Brain Products GmbH, Gilching, Germany). The second ECG (left clavicle – right hip) served as a back-up in case the signal quality of the first ECG (right clavicle – left hip) was too low for reliable R-peak or T-wave detection. In the current dataset, the first ECG signal was of high quality for all but one participant, for whom the second ECG was used in the analysis. R-peaks were identified using the Pan and Tompkins algorithm 62 , using code custom-written in MATLAB R2015b (The MathWorks Inc., Natick, MA, USA) combined with code modified from Sedghamiz, 2014 63 . The peaks of the P wave, Q wave, S wave, and T wave were identified using a custom-written code implementing the method proposed in Leutheuser et al, 2016 for detection of these four fiducial points 64 . The end of the T wave was determined using a custom-code implementing a trapezoidal area algorithm 65 , and was also adapted for detection of the beginning of the P wave. The pre-stimulus instantaneous heart rate was obtained by dividing 60 by the inter-beat interval (distance between two R-peaks) preceding RDM onset. Behavioural analysis The behavioural analyses were performed using custom-written scripts in MATLAB R2015b (The MathWorks Inc., Natick, MA, USA). For each subject, the disappearance probability during GFS trials in each of the three exercise intensity conditions was calculated by dividing the number of trials with at least one reported target disappearance by the total number of GFS trials performed. The reaction time for the catch trials was the average time taken after RDM onset/target removal to report the disappearance of the target. The average disappearance latencies for GFS trials were calculated by subtracting the reaction time for catch trials from the time at which the first disappearance was reported for each of the GFS trials with at least one reported target disappearance. The reappearance probability during GFS trials was calculated by dividing the number of trials with more than one reported target disappearance by the total number of GFS trials performed. Alpha amplitude analysis For the analysis of pre-stimulus alpha amplitudes, the data of the parieto-occipital electrodes O1, O2, Oz, POz, PO3, PO4, PO7, PO8, and Iz during the course of the whole trial were bandpass filtered at 8-12 Hz with a 4 th order Butterworth filter and subsequently Hilbert transformed. The pre-stimulus alpha amplitudes were obtained by taking the absolute values of the Hilbert transform, equivalent to the envelope of the filtered signal, in a time window spanning the second prior to the onset of the RDM stimulus (prestimulus window) pooled over the parieto-occipital electrodes. The data were baseline-corrected using a baseline window of 0.5 seconds prior to target onset. The electrodes were chosen based on previous research that investigated prestimulus alpha power occurring during GFS across the same set of electrodes 54 . The analysis for the other time windows (one second prior to target onset, one second following target onset, one second following RDM onset / post-RDM onset, the last second of RDM onset / 1 s post-RDM onset) was conducted in a similar fashion. The two seconds between target onset and RDM onset were split into two separate windows, instead of taking the whole time period as a pre-stimulus window to be able to distinguish between the response associated with target adaptation and the pre-stimulus preparatory processes prior to RDM onset, in other words, the neural factors that might lead to perceptual suppression. Similarly, the two seconds post-RDM onset were split into two separate windows (post-RDM onset and 1 s post-RDM onset), to distinguish between neural responses to the RDM onset and any subsequent perceptual response registration (which predominantly took place in the first second after RDM onset), and any post-response neural modulations. We also report differences in prestimulus pre-frontal alpha amplitude for the three exercise intensity conditions, calculated over the Fp1, Fp2, AF3, AF7, AFz, AF4, AF8 channels. For the comparison between exercise intensity conditions (levels: rest, low-resistance cycling, high-resistance cycling), we analysed the alpha amplitudes for all GFS trials for each of the three conditions retained after preprocessing. To determine each subject’s individual alpha frequency (IAF), we performed a Fast Fourier Transform (FFT) on the pre-stimulus time window for all parieto-occipital electrodes for each exercise intensity condition, analysing frequencies between 1 and 30 Hz with a resolution of 1 Hz. We then identified the peak frequency within the 8 to 12 Hz range for each subject. For topographical representation, we calculated the average absolute value of the 8–12 Hz filtered Hilbert transform during the second prior to RDM onset for each channel individually, then subtracted the rest condition values from those of the cycling conditions. For the two-factor comparison of target visibility (levels: visible, invisible) and exercise intensity, we analysed the alpha amplitudes separately for trials in which subjects had reported the target’s disappearance and trials in which that target was reported to remain visible, across each of the three exercise intensity conditions. Heartbeat-evoked potentials analysis Heartbeat-evoked potentials (HEPs) were computed on EEG signals locked to the T-peak of the ECG. Only trials with T-peaks occurring at least 300 ms after target onset up to 400 ms before RDM onset were chosen to avoid the HEPs from being contaminated by responses to the target or the RDM onset. The EEG signals were segmented from 1000 ms before the T-peak to 2000 ms after the T-peak, baseline corrected based on a -600 ms to -500 ms time window relative to the T-peak, then averaged to generate the prestimulus T-locked responses to heartbeats. Nine subjects had no trials that met the inclusion criteria for this analysis in the high-resistance condition. Additionally, eight subjects (not mutually exclusive from the nine previously mentioned) had heart rates in the high-resistance condition that were too high (R-to-R interval < 550 ms) to allow for a sufficient time window between heartbeats for statistical analysis. Consequently, the high-resistance condition was excluded from this analysis. One subject had no trials meeting the inclusion criteria for this analysis in the low-resistance condition and was therefore, also excluded from this analysis, resulting in a sample size of 29. For the HEP analysis, the significantly different heart rates between the two conditions exercise intensity conditions (rest vs low-resistance) were taken into account when selecting the statistical window. To account for variations in the timing of T-waves and P-waves and their respective cardiac field artifacts, the ends of the T-waves and the beginnings of the subsequent P-waves were carefully inspected on each trial. Based on this inspected, a 133 ms to 234 ms post-T-peak time window was chosen to ensure HEPs submitted for further statistical analysis are free from cardiac electrical artefacts. For the analysis of the effect of exercise intensity on HEPs, the selected window and all electrodes was submitted to a cluster-based permutation t -test 66 as implemented in the FieldTrip toolbox 61 . This test compared the HEP amplitudes between conditions (rest vs low-resistance), identifying clusters of significant differences across electrodes and time points. The Monte-Carlo method was used to generate the permutation distribution by randomly shuffling the condition labels 5000 times to calculate cluster-level statistics. Clusters were formed based on a t -statistic threshold of p < 0.05, with adjacent electrodes considered neighbours. Significant differences in clusters were identified if the cluster-level p value, corrected using the maximum cluster sum statistic, was below 0.05. To test for a 2 x 2 interaction effect of target visibility (visible vs invisible) and exercise intensity (rest vs low-resistance) on HEP amplitudes, a similar procedure was adopted, but a repeated measures permutation F -test was employed instead of a t­­ -test. This analysis compared the target visibility effect at rest (Rest(visible-invisible)) against the target visibility effect at low-resistance (Low-resistance(visible-invisible)) to assess whether the target visibility effect differed significantly between exercise intensity conditions. For any significant clusters identified by the cluster-based permutation tests, the average HEP amplitudes over the electrodes and time points within the cluster were extracted. These average amplitudes were then submitted to paired samples t -tests to further validate the observed differences between conditions. Control analyses for possible effects of cardiovascular artifacts HEPs represent neural responses to cardiac signals but can also include cardiac field artifacts and pulse-related artifacts 67 . One commonly used approach to mitigate any associated volume conduction effects of these artifacts is an independent component analysis (ICA). However, this approach has been criticised for its limited ability to completely eliminate the cardiac field artifacts and for the potential risk of removing relevant task-related signals 51 . To ensure that the observed effects in HEP amplitudes were not due to differences in cardiac electrical activity directly affecting EEG data by volume conduction, we submitted the mean ECG amplitudes to the same cluster-based permutation tests as the observed effects. Several studies have also controlled for heart rate when differences in HEPs were observed 48,49 . By design of using exercise as an intervention, the rest and low-resistance conditions have significantly different heart rates. We thus fitted a linear mixed-effect model (LME) to investigate the effects of exercise intensity conditions and heart rate on HEP amplitude. The model examined the main effects of exercise intensity condition and heart rate on HEP amplitude, with condition and heart rate as fixed effects and subject included as a random intercept to account for repeated measures within participants. Assumptions of linearity, homoscedasticity, and no multicollinearity were evaluated prior to model fitting, while the normality of residuals was assessed post hoc . Model fitting and parameter estimation were conducted using maximum likelihood estimation in MATLAB R2015b (The MathWorks Inc., Natick, MA, USA). Model fit was assessed using the Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and log-likelihood values, with statistical significance set at α = 0.05. Statistical analyses Statistical analyses were conducted in SPSS Statistics for Windows, Version 27.0 (IBM Corp, Armonk, NY, USA). Data were tested for the normality assumption using the Shapiro-Wilk test. When testing for the effects of exercise intensity (3 levels: rest, low-resistance cycling, high-resistance cycling), data that met the normality assumption were analysed using the repeated-measures ANOVA (RM-ANOVA), followed by Bonferroni-corrected post-hoc pairwise comparisons when significant effects were found. When testing for the effects of target visibility and exercise intensity, when normality was satisfied, a 2 (factor, target visibility; levels: visible, invisible) x 3 (factor, exercise intensity; levels: rest, low-resistance cycling, high-resistance cycling) repeated-measures ANOVA was applied. Mauchly’s test was used to assess sphericity, and if this assumption was violated, the Greenhouse-Geisser correction was applied. If the test revealed significant effects, post-hoc Bonferroni-corrected pairwise comparisons were performed. Data that did not satisfy the normality assumption were analysed using the Friedman test to assess differences across conditions. When the Friedman test indicated a significant effect, post-hoc pairwise comparisons were conducted using the Wilcoxon signed-rank test to identify specific differences between conditions, with a Bonferroni correction applied for multiple comparisons. For the analysis of the effects of exercise intensity and target visibility on alpha amplitudes, a Bonferroni-Holm correction was applied due to the larger number of comparisons in this analysis. Partial eta-squared (\(\eta_{p}^{2}\)), Cohen’s d , Kendall’s W , and rank-biserial correlation ( r ) were calculated as the effect size for the repeated-measures ANOVAs, t -tests, Friedman tests, and Wilcoxon signed-rank tests, respectively. For comparisons against zero, a one-sample t -test was used for data meeting the normality assumption, while a Wilcoxon signed-rank test was applied for data that did not meet this assumption. For correlation analyses, Pearson’s correlation was used for normally distributed data, while Kendall’s Tau-b was applied for data that did not meet the normality assumption. Results Behavioural analysis Subjects performed a Generalized Flash Suppression (GFS) task under three exercise intensity conditions: at rest, and during low-resistance or high-resistance cycling. In the GFS task, a salient target stimulus is typically completely suppressed, i.e. rendered subjectively invisible, after the onset of a random dot motion (RDM) stimulus (see section 2.2 - Stimuli and task ). We tested if there was an effect of exercise on the probability of target suppression during GFS trials ( Figure 2A ). The RM-ANOVA revealed an effect of exercise intensity on disappearance (i.e. target suppression) probability ( F (2, 58) = 6.40, p = 3.08\(\times\)10 -3 , \(\eta_{p}^{2}\) = 0.18). Post-hoc Bonferroni-corrected pairwise comparison showed that the disappearance probability was lower for the low-resistance (mean \(\pm\)SEM: 0.43 \(\pm\) 0.04; p = 0.02) and high-resistance (mean\(\pm\) SEM: 0.41 \(\pm\) 0.04; p = 0.01) cycling conditions compared with rest (mean \(\pm\) SEM: 0.50 \(\pm\) 0.03), but there was no difference between the two cycling conditions ( p = 1.00). On average, this corresponded to a 14.99% decrease in disappearance probability for the low-resistance condition (one-sample t- test, t (29) = -2.92, p = 6.78\(\times\)10 -3 , Cohen’s d = -0.53) and 18.45% decrease for the high-resistance condition (one-sample t- test, t (29) = -3.15, p = 3.74\(\times\)10 -3 , Cohen’s d = -0.58) compared to rest ( Figure 2B ). ( A ) Violin plots showing the mean disappearance probabilities during GFS performed at rest (blue), low-resistance (orange), and high-resistance (green) cycling. Statistical significance between the conditions as assessed by post hoc Bonferroni-corrected pairwise comparisons is indicated at the p \(\leq\) 0.05 (*) level. ( B ) Violin plots showing the percent change in disappearance probability for low-resistance (orange) and high-resistance (green) cycling compared to rest. Statistical significance as assessed by one-sample t -tests is indicated at the p \(\leq\) 0.01 (**) level. For both panels, vertical grey boxes indicate the interquartile range (IQR), with grey whiskers indicating 1.5 x IQR, surrounded on each side by the kernel density estimation in the colour corresponding to each group. Horizontal lines in the colour corresponding to each group denote means and white dots indicate medians. Each dot in a given condition represents the average value of a single subject. N = 30 for all plots. No significant differences between the three exercise intensity conditions were observed in the percentage of correct reports during both RDM OFF trials (Friedman test, \(\chi^{2}\)(2) = 0.19, p = 0.91, Kendall’s W = 3.11\(\times\)10 -3 ; overall mean \(\pm\) SD: 96.52\(\pm\) 4.98%; median: 100%) and catch trials (Friedman test,\(\chi^{2}\)(2) = 3.08, p = 0.22, Kendall’s W = 0.05; overall mean \(\pm\) SD: 97.32 \(\pm\) 4.90%; median: 100%). The high detection rates across these control trials suggest subjects’ percept detection and reporting abilities remained unimpaired across the exercise intensity conditions. Alpha amplitude analysis Given that cycling modulates alpha activity (8-12 Hz) 5,6 and pre-stimulus parieto-occipital alpha amplitudes are predictive of perceptual suppression 54 , we analysed the effect of exercise intensity on parieto-occipital alpha amplitudes during the second preceding RDM onset. There was a statistically significant difference in the mean parieto-occipital alpha amplitude over this time window ( Figure 3 and Figure 4A ) between exercise intensity conditions (Friedman test,\(\chi^{2}\)(2) = 16.80, p = 2.25\(\times\)10 -4 , Kendall’s W = 0.29). Parieto-occipital alpha amplitudes were lower during cycling compared to rest (Bonferroni-corrected Wilcoxon signed-rank tests, low-resistance: Z = -3.86, p = 3.45\(\times\)10 -4 , r = 0.50; high-resistance: Z = -2.89, p = 0.01, r = 0.37), but there was no difference between the two cycling conditions ( Z = -0.85, p = 1.18, r = 0.11). Average time courses of alpha band (8 – 12 Hz) amplitudes for GFS performed during ( A ) low-resistance (orange) and ( B ) high-resistance (green) cycling compared to rest (blue) ( N = 30). The shaded areas denote the SEM for each time course. The zero mark denotes the onset of the RDM stimulus, after which the target was potentially perceptually suppressed. The data represent the mean of all parieto-occipital electrodes. Statistical significance between the conditions for different time windows as assessed by Bonferroni-corrected Wilcoxon signed-rank tests is indicated at the p \(\leq\) 0.05 (*), p \(\leq\) 0.001 (***), and p \(\leq\) 0.0001 (****) levels. We also found differences in parieto-occipital alpha amplitude due to exercise intensity in other time windows ( Figure 3 ), specifically, during the first second of RDM presentation (post-RDM onset, Friedman test, \(\chi^{2}\)(2) = 29.07, p = 4.88\(\times\)10 -7 , Kendall’s W = 0.48) and the following half-second (1 s post-RDM onset, Friedman test,\(\chi^{2}\)(2) = 26.87, p = 1.47\(\times\)10 -6 , Kendall’s W = 0.45). Post-hoc Bonferroni-corrected Wilcoxon signed-rank tests revealed a significant decrease in parieto-occipital alpha amplitude during cycling as compared to rest in both these time windows (post-RDM onset results: low-resistance, Z = -3.92, p = 2.68\(\times\)10 -4 , r = 0.50; high-resistance, Z = -3.98, p = 2.07\(\times\)10 -4 , r = 0.51; 1 s post-RDM onset results: low-resistance, Z = -3.63, p = 8.49\(\times\)10 -4 , r = 0.47; high-resistance, Z = -4.19, p = 8.53\(\times\)10 -4 , r = 0.54). Again, no significant differences were observed between the cycling conditions (post-RDM onset results: Z = -0.09, p = 2.78, r = 0.01; 1 s post-RDM onset results: Z = -0.92, p = 0.36, r = 0.12). ( A ) Violin plots showing the mean pre-stimulus parieto-occipital alpha during GFS performed at rest (blue), low-resistance (orange), and high-resistance (green) cycling. Vertical grey boxes indicate the interquartile range (IQR), with grey whiskers indicating 1.5 \(\times\) IQR, surrounded on each side by the kernel density estimation in the colour corresponding to each group. Horizontal lines in the colour corresponding to each group denote means and white dots indicate medians. Each dot in a given condition represents the average value of a single subject. Statistical significance between conditions as assessed by Bonferroni-corrected Wilcoxon signed-rank tests is indicated at the p \(\leq\) 0.05 (*) and p \(\leq\) 0.001 (***) levels. Power spectra of the second preceding the RDM stimulus for GFS performed at rest and during ( B ) low-resistance and ( C ) high-resistance cycling across subjects. Topography of the ( D ) low-resistance \(–\) rest and ( E ) high-resistance \(–\) rest differences in 8-12 Hz amplitude in the second prior to RDM onset. N = 30 for all plots. The power spectra of the second prior to RDM onset revealed that the peak individual alpha frequency (IAF) did not differ across the three exercise intensity conditions (Friedman test, \(\chi^{2}\)(2) = 5.26, p = 0.07, Kendall’s W = 0.09) and the mean IAF across subjects was 9.62 Hz \(\pm\) SD 1.05 Hz. However, the cycling conditions led to a downshift of the EEG power spectrum compared to rest, and an analysis of the power at peak IAF ( Figure 4B and C ) revealed a difference between the three conditions (Friedman test, \(\chi^{2}\)(2) = 15.27, p = 4.84\(\times\)10 -4 , Kendall’s W = 0.25). Corroborating the pre-stimulus cycling-induced decrease in alpha amplitude, decreased power was observed for low-resistance (Bonferroni-corrected Wilcoxon signed-rank test, Z = -3.67, p = 7.24\(\times\)10 -4 , r = 0.47) and high-resistance ( Z = -2.77, p = 0.02, r = 0.36) cycling compared to rest, but there was no difference in power between cycling conditions ( Z = -0.36, p = 2.16, r = 0.05). The topographies of the cycling – rest difference in alpha amplitude ( Figure 4D ) suggested that the decrease in alpha activity was most prominent in the parieto-occipital cortex. Correlations as assessed by Kendall’s Tau-b between percent change in parieto-occipital alpha amplitude from rest and percent change in disappearance probability from rest for ( A ) low-resistance and ( B ) high-resistance cycling. N = 30 for all plots. In addition to parieto-occipital alpha amplitude, we examined pre-stimulus pre-frontal alpha amplitude, as exercise has been reported to elicit prominent effects in the alpha band in anterior regions 6 . A significant effect of exercise intensity on pre-stimulus pre-frontal alpha was observed (Friedman test,\(\chi^{2}\)(2) = 18.07, p = 1.19\(\times\)10 -4 , Kendall’s W = 0.30). Post-hoc Bonferroni-corrected Wilcoxon signed-rank tests revealed that both low-resistance (Z = -4.10, p = 1.22\(\times\)10 -4 , r = 0.75) and high-resistance (Z = -3.65, p = 7.84\(\times\)10 -4 , r = 0.67) cycling conditions resulted in significantly lower pre-frontal alpha amplitudes compared to rest. However, no significant difference was found between the two cycling conditions (Z = -0.61, p = 1.63, r = 0.11). We also examined whether changes in pre-frontal alpha amplitude correlated with changes in disappearance probability. No significant correlation was observed between the percentage change in pre-stimulus pre-frontal alpha amplitude from rest and the percentage change in disappearance probability for either low-resistance (Kendall’s Tau-b,\(\tau_{b}\)(30) = -0.05, p = 0.78) or high-resistance cycling (\(\tau_{b}\)(30) = -0.08, p = 0.56). In an analysis of the relationship between cycling-related changes in alpha amplitude on changes in disappearance probability ( Figure 5 ), we observed no statistically significant correlations for neither low-resistance and high-resistance cycling conditions (Kendall’s Tau-b, all p s \(>\) 0.05). Based on previous work on GFS performed at rest 54 , we hypothesized that pre-stimulus parieto-occipital alpha would be lower when the target becomes perceptually suppressed compared to when it remains visible. To investigate this, we assessed whether pre-stimulus, parieto-occipital alpha activity reflects the subjective visibility of the target in GFS trials ( Figure 6 ) and examined how this relationship interacted with the effect of cycling. A Friedman test revealed a statistically significant difference in pre-stimulus, parieto-occipital alpha activity across the exercise intensity and target visibility conditions (\(\chi^{2}\)(5) = 42.74, p = 4.17\(\times\)10 -8 , Kendall’s W = 0.29). In the post hoc analysis, we performed nine pairwise comparisons with the Bonferroni-Holm-corrected Wilcoxon signed-rank test ( Figure 6D ), with three comparisons testing for the effect of exercise intensity on pre-stimulus alpha amplitudes and six comparisons testing for the effect of pre-stimulus alpha amplitudes on target visibility. This analysis suggests alpha amplitude-linked effects of both target visibility and exercise intensity. Violin plots in Figure 6E show that for most subjects, pre-stimulus alpha amplitude was consistently higher for trials where the target remained visible compared to when it was rendered invisible across all exercise intensity conditions, in line with the findings in Poland et al (2021) 54 . However, the target visibility-related effects do not survive the correction for multiple comparisons (rest: Z = -2.03, p = 0.17; low-resistance: Z = -2.07, p = 0.19; high-resistance: Z = -1.98, p = 0.14; Uncorrected p values reported in Supplementary Information , Table S1 ). As for the cycling-related effects on pre-stimulus alpha, we observed significant reductions in alpha amplitude during visible trials for both low-resistance ( Z = -3.82, p = 1.09\(\times\)10 -3 , r = 0.49) and high-resistance cycling ( Z = -2.95, p = 0.02, r = 0.38) compared to rest, with no significant difference between the two cycling conditions ( Z = -1.41, p = 0.32, r = 0.18). Similarly, in the invisible trials, a significant reduction was found between rest and low-resistance cycling ( Z = -3.90, p = 8.74\(\times\)10 -4 , r = 0.50), but not between rest and high-resistance cycling ( Z = -2.58, p = 0.06, r = 0.33) or between the two cycling conditions ( Z = -1.08, p = 0.28, r = 0.14). These results suggest that while cycling decreases pre-stimulus alpha amplitude, this exercise-induced modulation is not specifically linked to perceptual suppression. Average time courses of alpha band (8 – 12 Hz) amplitudes for GFS trials where the target remained visible and trials where the target was rendered invisible when the task was performed during ( A ) rest, ( B ) low-resistance and ( C ) high-resistance cycling. SEM plotted for each time course in corresponding colour. The zero mark denotes the onset of the RDM stimulus, after which the target was potentially perceptually suppressed. Data represent the mean of all parieto-occipital electrodes. Statistical significance between the conditions for different time windows as assessed by Wilcoxon signed-rank tests indicated at the p \(\leq\) 0.05 (*) level. ( D ) Bonferroni-Holm-corrected Wilcoxon signed-rank test p -values for pairwise comparisons reflecting the effect of target visibility and exercise intensity on parieto-occipital alpha amplitude in the second prior to RDM onset (Rest – R, Low-resistance – L, High-resistance – H, Visible – vis, Invisible - inv). ( E ) Violin plots showing the visible – invisible difference in mean pre-stimulus alpha during GFS performed at rest (blue), and during low-resistance (orange), and high-resistance (green) cycling. Vertical grey boxes indicate the interquartile range (IQR), with grey whiskers indicating 1.5 \(\times\) IQR, surrounded on each side by the kernel density estimation in the colour corresponding to each group. Horizontal lines in the colour corresponding to each group denote means and white dots indicate medians. Each dot in a given condition represents the average value of a single subject. N = 30 for all plots. Heart rate analysis To investigate how the pre-stimulus instantaneous heart rate interacts with exercise intensity and conscious visual perception of the target, we conducted a two-way repeated measures ANOVA with two factors: exercise intensity (levels: rest, low-resistance cycling, high-resistance cycling) and target visibility (levels: visible, invisible). The analysis revealed a significant main effect of exercise intensity (Greenhouse-Geisser corrected; F (1.25, 36.26) = 269.24, p = 1.09\(\times\)10 -19 , \(\eta_{p}^{2}\) = 0.90), confirming the efficacy of the exercise intervention. Post-hoc Bonferroni-corrected pairwise comparison indicated that both low-resistance and high-resistance cycling elicited significantly higher heart rates compared to rest ( Figure 7A ; low-resistance: 10.81 bpm difference, p = 2.78\(\times\)10 -16 ; high-resistance: 28.59 bpm difference, p = 4.53\(\times\)10 -17 ). Additionally, high-resistance cycling resulted in a significantly higher heart rate than low-resistance cycling (17.77 bpm difference, p = 3.11\(\times\)10 -13 ). On average, the heart rate increased by 15.56% during low-resistance cycling compared to rest, and by 41.32% during high-resistance cycling compared to rest. However, there was no significant main effect of target visibility: the instantaneous pre-stimulus heart rate did not differ significantly between trials in which the target remained visible and those in which it disappeared ( F (1, 29) = 0.99, p = 0.33,\(\eta_{p}^{2}\) = 0.03). Furthermore, no significant interaction between exercise intensity and target visibility was observed ( F (2, 58) = 0.47, p = 0.63, \(\eta_{p}^{2}\) = 0.02). We investigated whether these heart rate changes were associated with the cycling-related changes in disappearance probability and pre-stimulus alpha amplitude. For the relationship between heart rate and disappearance probability, there were no statistically significant correlations for either low-resistance or high-resistance cycling ( Figure 7B and C ; Kendall’s tau-b, all p s > 0.05). Similarly, for the relationship between heart rate and alpha amplitude, no statistically significant correlations emerged for either condition ( Figure 7D and E ; Kendall’s tau-b, all p s > 0.05). We also assessed phasic heart rate changes across the course of a trial, and found a deceleration in heart rate over the trial duration, which occurred irrespective of target visibility ( Supplementary Information , Figure S1 and Table S2 ). This deceleration aligns with the well-characterized phenomenon of cardiac deceleration following a warning stimulus, which reflects anticipation of an upcoming stimulus or response registration 20–22,34–40 . Interestingly, in our paradigm, this phenomenon emerged despite the absence of an explicit warning stimulus, likely driven instead by the presence of stereotyped, predictable events that may have acted as implicit cues for anticipation. Furthermore, this deceleration appears to be robust, persisting even under the cycling conditions. ( A ) Violin plots showing the pre-stimulus instantaneous heart rate during GFS performed at rest (blue), low-resistance (orange), and high-resistance (green) cycling. Vertical grey boxes indicate the interquartile range (IQR), with grey whiskers indicating 1.5 x IQR, surrounded on each side by the kernel density estimation in the colour corresponding to each group. Horizontal lines in the colour corresponding to each group denote means and white dots indicate medians. Each dot in a given condition represents the average value of a single subject. Statistical significance between the conditions as assessed by a post hoc analysis with a Bonferroni adjustment is indicated at the p \(\leq\) 0.0001 (****) level. Correlations as assessed by Kendall’s Tau-b between percent change in heart rate from rest and percent change in disappearance probability from rest for ( B ) low-resistance and ( C ) high-resistance cycling. Correlations between percent change in heart rate from rest and percent change in alpha amplitude from rest for ( D ) low-resistance and ( E ) high-resistance cycling. All correlations are not significant. N = 30 for all plots. Heartbeat-evoked potentials analysis To explore if and how neural processing of heartbeats is altered during exercise, we analysed the effect of cycling on heartbeat-evoked potential (HEP) amplitudes. EEG data locked to the T-peak preceding RDM onset was averaged for the rest and low-resistance cycling conditions. The high-resistance cycling condition was excluded from this analysis because the heart rates were too high to allow for a sufficient time window between heartbeats for statistical analysis. For the other two conditions, a 133 to 234 ms post-T-peak time window, chosen to avoid overlap with the end of the previous T wave and the beginning of the subsequent P wave, ensuring HEPs are free from cardiac artifacts, was submitted to a cluster-based permutation t -test. Pre-stimulus HEPs significantly differed between rest and low-resistance cycling ( Figure 8A , B , C ) over central, parietal, and occipital electrodes in a 164 to 203 ms post-T-peak time window (cluster-level statistic = 291.24, p = 0.03, N = 29; Supplementary Information , Figure S2 ). The HEP amplitudes during low-resistance cycling were lower than at rest (paired samples t -test, t (28) = -2.19, p = 0.04, Cohen’s d = 0.41; Figure 8D ). The control for volume conduction, i.e. comparison of the ECG waveforms for rest and low-resistance in the same time window as the observed effect, revealed no significant differences ( Supplementary Information , Figure S3 ). Thus, the observed differences in HEP amplitudes between the two conditions cannot be attributed to differences in cardiac electrical activity. The linear mixed-effects model ( Supplementary Information , Table S2 ) used to investigate the effects of exercise intensity conditions on HEP amplitude, while controlling for heart rate, revealed no significant difference in HEP amplitude between rest and low-resistance cycling ( b = 0.26, p = 0.26). Heart rate showed a marginal negative association with HEP amplitude ( b = -0.02, p = 0.09), suggesting a potential trend toward reduced HEP amplitudes at higher heart rates. Random intercepts for subject (\(\sigma\) = 0.30) accounted for individual variability, and the residual standard deviation was 0.77. Model fit statistics included AIC = 151.68 and BIC = 161.98. These results suggest that heart rate is a likely contributor to the observed difference in HEP amplitudes between conditions. ( A ) Topographical map of heartbeat-evoked potential (HEP) amplitude difference between low-resistance and rest conditions in the 164 to 203 ms post-T-peak time window during which a statistically significant difference was observed. White labels indicate the channels contributing to the significant cluster. ( B ) Pre-stimulus HEPs for the two conditions averaged across the cluster. The shaded areas denote the SEM for each time course. The signal contaminated by T-wave and P-wave cardiac artifacts appears in lighter colour. The grey bar highlights the time window in which a significant difference was observed as the p \(\leq\) 0.05 (*) level. ( C ) Violin plots of the HEP amplitude averaged across the cluster for rest (blue) and low-resistance cycling (orange). Vertical grey boxes indicate the interquartile range (IQR), with grey whiskers indicating 1.5 x IQR, surrounded on each side by the kernel density estimation in the colour corresponding to each group. Horizontal lines in the colour corresponding to each group denote means and white dots indicate medians. Each dot in a given condition represents the average value of a single subject. For both ( B ) and ( C ) Statistical significance between the conditions as assessed by paired samples t -test at the p \(\leq\) 0.05 (*) level. ( D ) Difference from rest in HEP amplitudes for low-resistance cycling. Statistical significance as assessed by one-sample t -test is indicated at the p \(\leq\) 0.05 (*) level. Correlations as assessed by Pearson correlation between ( E ) change in HEP amplitude and percent change in disappearance probability and ( F ) change in heart rate and chance in HEP amplitude for cycling from rest. Neither correlation is significant. N = 29 for all plots. We investigated whether these modulations in HEPs were associated with cycling-related changes in disappearance probability ( Figure 8E ) and found no statistically significant correlation (Pearson’s correlation, r(29) = -0.17, p = 0.39). We also investigated the relationship between the cycling-related change in heart rate and the change in HEP amplitude ( Figure 8F ) and no statistically significant correlation was found (Pearson’s correlation, r(29) = 0.05, p = 0.82). Additionally, HEP amplitude has been linked to conscious perception of sensory stimuli at threshold 37,52 , suggesting a role of interoceptive processing in the perception of external stimuli. To investigate this, we employed a cluster-based permutation F -test to assess the 2×2 interaction effect of target visibility (levels: visible, invisible) and exercise intensity (levels: rest, low-resistance) on HEP amplitudes. This analysis revealed two positive spatio-temporal clusters, however, neither cluster reached statistical significance (smallest cluster-level p = 0.32). Instead, we assessed whether pre-stimulus HEP amplitude reflected the subjective visibility of the target in GFS trials ( Figure 9 ) within the same time window as the exercise intensity effect described above, as well as its interaction with the cycling effect. A Friedman test revealed a statistically significant difference in HEP amplitudes across the exercise intensity and target visibility conditions (\(\chi^{2}\)(3) = 8.63, p = 0.04, Kendall’s W = 0.10). In the post hoc analysis, we performed four pairwise comparisons with the Bonferroni-corrected Wilcoxon signed-rank test ( Figure 9C ), with two comparisons testing for the effect of exercise intensity on pre-stimulus HEP amplitudes and two comparisons testing for the effect of pre-stimulus HEP amplitudes on target visibility. However, neither the effect of target visibility nor the effect of cycling survived the correction for multiple comparisons (all p s > 0.05; Uncorrected p values reported in Supplementary Information , Table S3 ). The difference between HEP amplitudes for trials in which the target remains visible and the trials in which it is perceptually suppressed ( Figure 9D ) is not significant for rest ( t (28) = -1.53, p = 0.14, Cohen’s d = -0.29) or low-resistance cycling ( t (28) = -0.38, p = 0.71, Cohen’s d = -0.07). Pre-stimulus HEPs averaged across the significant cluster for the trials where the target remained visible and was rendered invisible during GFS, shown for ( A ) rest and ( B ) low-resistance cycling conditions. ( C ) Bonferroni-corrected Wilcoxon signed-rank test p -values for pairwise comparisons reflecting the effect of target visibility and exercise on pre-stimulus HEP amplitude (Rest – R, Low-resistance – L, Visible – vis, Invisible - inv). ( D ) Violin plots showing the visible – invisible difference in mean pre-stimulus HEP amplitude (from the same time window as the exercise intensity analysis) during GFS performed at rest (blue) and during low-resistance (orange) cycling. Vertical grey boxes indicate the interquartile range (IQR), with grey whiskers indicating 1.5 \(\times\) IQR, surrounded on each side by the kernel density estimation in the colour corresponding to each group. Horizontal lines in the colour corresponding to each group denote means and white dots indicate medians. Each dot in a given condition represents the average value of a single subject. N = 29 for all plots. Discussion In the current study, we investigated behavioural, neural, and cardiovascular responses to Generalized Flash Suppression (GFS) performed during light-intensity cycling. We found that the disappearance probability, i.e. the rate of perceptual suppression, decreased during cycling. In parallel, pre-stimulus parieto-occipital alpha activity were lower during cycling compared to rest. When comparing pre-stimulus alpha modulations between visible and invisible trials, we observed that parieto-occipital alpha amplitudes were lower prior to target disappearance compared to when the target remained visible, consistently across all three exercise intensity conditions. While this difference did not reach significance after correction for multiple comparisons, the direction of the effect aligns with our previous EEG GFS study 54 . These results, along with the absence of a correlation between cycling-related modulation of alpha amplitude and disappearance probability, suggest that while pre-stimulus alpha amplitudes index target visibility, they do not account for the decreased rate of perceptual suppression during cycling. Modulations of alpha activity have been shown to reflect fluctuations in attentional states, with the processing of attended stimuli being linked to reduced alpha activity compared to unattended stimuli 30–32 . Such reductions in alpha activity are thought to indicate heightened neural excitability 25–29 and are linked to the suppression of distracting stimuli 30,68–70 . In the previous GFS study linking alpha decreases to target disappearance at rest, it was proposed that these functions of alpha were responsible for enhancing the processing of the surrounding motion stimulus, thereby increasing its effectiveness in supressing GFS targets 54 . In the current paradigm, we cannot distinguish between neural modulations associated with target and surround, which would be of interest to investigate in future studies. Understanding whether the observed alpha reductions during cycling reflect attentional shifts specific to the target, the surrounding stimulus, or a combination of both could provide further insights into the neural mechanisms underlying the reduced rate of perceptual suppression in exercise contexts. While the observed decreases in alpha activity indicate increased visual attention 30–32 , they may not be solely attributable to target or surround processing. Instead, these reductions could reflect the dual-task demands of dividing external attention between perceptual and motor task performance. Similar alpha modulations have been observed during walking, where reductions in parietal alpha are thought to reflect sensorimotor integration during visually guided walking in a virtual reality paradigm 71 . Walking has been linked to more pronounced alpha power decreases than cycling, potentially reflecting the unique somatosensory and proprioceptive demands of gait 13 . Correspondingly, unique patterns of visual processing have been found associated with walking and cycling. For example, studies employing orientation discrimination paradigms report different outcomes during walking and cycling. Concurrent decreases in both pre-stimulus alpha power and orientation discrimination accuracy are reported during natural walking (as opposed to stationary walking on a treadmill), with the changes in alpha suggesting a change in attentional state linked to visual awareness 72 . Meanwhile during cycling, increases in parieto-occipital alpha relative to rest have been associated with a decrease in orientation discrimination accuracy 9 . These differing findings suggest that changes in alpha activity may vary depending on the type of exercise, potentially leading to different perceptual outcomes based on the sensory and attentional demands associated with each exercise. In addition to neural and behavioural effects, our study examined exercise-mediated cardiac influences on neural activity and perceptual outcomes in GFS. As expected, there was an increase in heart rate associated with the physical activity of cycling. However, this increase in heart rate did not correlate with the observed decreases in disappearance probability or pre-stimulus alpha activity, nor did it differ significantly between trials in which the target remained visible and those in which it disappeared. The decrease in alpha observed in parallel to the increase in heart rate, argues against the baroreceptor hypothesis theory of cardiovascular influences on cortical excitability during exercise 5,6 . These results suggest that the sustained changes in heart rate due to cycling do not reflect attentional and perceptual processes, unlike phasic changes occurring in anticipation of a stimulus or following response registration 20–22,34–40 , which we also observed in the current study. Instead, these induced changes in heart rate due to cycling are likely indicative of sympathetic dominance and a tendency for the system to be biased towards action rather than perception 42 . This sustained sympathetic dominance may overshadow more subtle interoceptive effects that could contribute to perceptual suppression, masking their influence during exercise. In addition to changes in heart rate and alpha activity, we also observed cycling-related differences in HEPs, reflecting a change in the neural responses to cardiac signals during exercise. Increased HEP amplitudes have been linked to enhanced interoceptive attention, reflecting both explicit attention 50,51,73–76 to and unconscious processing of heartbeats 67,77 . We observed a decrease in HEP amplitude during cycling compared to rest, suggesting that interoceptive processing is attenuated during exercise, potentially reflecting an attentional shift away from interoceptive information and towards exteroceptive information required for simultaneously performing a perceptual and motor task. Our results also indicate that the difference in heart rate between the rest and low-resistance cycling conditions contributes to the observed difference in HEP amplitudes between them. Taken together, these results imply that increased heart rate and associated increases in baroreceptor firing, resulted in a decrease in interoceptive processing. Furthermore, these finding argue against the idea that cardiac parameters have a limited influence on HEPs 48,49,78 , suggesting that interpreting HEPs may require further careful consideration. Variability in HEPs is proposed to have two possible nonexclusive origins: changes in neural dynamics or changes in cardiac physiology 37 , and here we have tentative evidence to suggest that cardiac physiology does indeed affect HEPs. Furthermore, in our study, the HEPs did not predict perceptual suppression in GFS at rest or during cycling, nor did cycling-induced changes in HEPs correlate with changes in disappearance probability. These results suggest that while HEPs are sensitive to changes in physiological state, their role in mediating perceptual outcomes may be limited, at least in the context of GFS. This is in contrast with preliminary evidence from previous research reporting that increased HEP amplitudes were associated with increased detection of a target in a visual threshold paradigm 37 . The absence of such a link in our study suggests that the influence of interoceptive processing as indexed by HEPs on perceptual processing may vary depending on the perceptual task, highlighting the need for further investigation into the contribution of HEPs to conscious perception. The observed exercise-linked reductions in perceptual suppression, alpha activity and HEP amplitude highlight a complex interplay between interoception and exteroception during physical activity. While exercise appears to enhance exteroceptive processing, as reflected in reduced perceptual suppression and reduced alpha amplitudes, it simultaneously attenuates cardiac interoceptive processing, as indexed by reduced HEPs. These findings support the idea of a dynamic trade-off between interoceptive and exteroceptive domains, modulated by task demands and physiological state. Author contributions / CRediT statement Aishwarya Bhonsle : Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Software; Visualization; Writing – original draft; Writing – review & editing Melanie Wilke : Conceptualization; Methodology; Resources; Supervision; Writing – review & editing, Funding acquisition Funding information This research was supported by the Hermann and Lilly Schilling Foundation, the Else-Kröner-Fresenius Foundation and the DFG GRK 2824: ‘Heart and Brain’. Conflict of interest statement The authors declare no conflicts of interest. Data availability statement Data available on request. Acknowledgements We would like to thank Greta Wippich for the illustration of the experimental setup; Alina Seidel for help with data collection; Severin Heumüller for providing technical support; the UMG Statistische Betreuung team for advice on the statistical analyses; Carsten Schmidt-Samoa, Shirin Mahdavi, Ulrich Parlitz and Roberto Goya-Maldonado for helpful discussions. References 1. 1. Lambourne, K. & Tomporowski, P. The Effect of Exercise-Induced Arousal on Cognitive Task Performance: A Meta-Regression Analysis. Brain Res 1341 , 12–24 (2010).2. Chang, Y. K., Labban, J. D., Gapin, J. I. & Etnier, J. L. The Effects of Acute Exercise on Cognitive Performance: A Meta-Analysis. Brain Research 1453 , 87–101 (2012).3. Cantelon, J. A. & Giles, G. E. A Review of Cognitive Changes During Acute Aerobic Exercise. Front. Psychol. 12 , (2021).4. Garrett, J., Chak, C., Bullock, T. & Giesbrecht, B. A Systematic Review and Bayesian Meta-Analysis Provide Evidence for an Effect of Acute Physical Activity on Cognition in Young Adults. Commun Psychol 2 , 82 (2024).5. Crabbe, J. B. & Dishman, R. K. Brain Electrocortical Activity During and After Exercise: A Quantitative Synthesis. Psychophysiology 41 , 563–574 (2004).6. Hosang, L., Mouchlianitis, E., Guérin, S. M. R. & Karageorghis, C. I. Effects of Exercise on Electroencephalography-Recorded Neural Oscillations: A Systematic Review. International Review of Sport and Exercise Psychology 0 , 1–54 (2022).7. Heinonen, I. et al. Organ-Specific Physiological Responses to Acute Physical Exercise and Long-Term Training in Humans. Physiology 29 , 421–436 (2014).8. Cao, L. & Händel, B. Walking Enhances Peripheral Visual Processing in Humans. PLoS Biol 17 , e3000511 (2019).9. Bullock, T., Elliott, J. C., Serences, J. T. & Giesbrecht, B. Acute Exercise Modulates Feature-selective Responses in Human Cortex. J Cogn Neurosci 29 , 605–618 (2017).10. Bullock, T., Cecotti, H. & Giesbrecht, B. Multiple Stages of Information Processing Are Modulated During Acute Bouts of Exercise. Neuroscience 307 , 138–150 (2015).11. De Sanctis, P., Butler, J. S., Malcolm, B. R. & Foxe, J. J. Recalibration of Inhibitory Control Systems During Walking-Related Dual-Task Interference: A Mobile Brain-Body Imaging (MOBI) Study. NeuroImage 94 , 55–64 (2014).12. Ciria, L. F. et al. Oscillatory Brain Activity During Acute Exercise: Tonic and Transient Neural Response to an Oddball Task. Psychophysiology 56 , e13326 (2019).13. Storzer, L. et al. Bicycling and Walking are Associated with Different Cortical Oscillatory Dynamics. Front. Hum. Neurosci. 10 , (2016).14. Hottenrott, K., Taubert, M. & Gronwald, T. Cortical Brain Activity Is Influenced by Cadence in Cyclists. Open Sports Sci J 6 , 9–14 (2013).15. Enders, H. et al. Changes in Cortical Activity Measured with EEG During a High-Intensity Cycling Exercise. Journal of Neurophysiology 115 , 379–388 (2016).16. Bailey, S. P., Hall, E. E., Folger, S. E. & Miller, P. C. Changes in EEG During Graded Exercise on a Recumbent Cycle Ergometer. Journal of Sports Science & Medicine 7 , 505 (2008).17. Fumoto, M. et al. Ventral Prefrontal Cortex and Serotonergic System Activation During Pedaling Exercise Induces Negative Mood Improvement and Increased Alpha Band in EEG. Behavioural Brain Research 213 , 1–9 (2010).18. Garber, C. E. et al. Quantity and Quality of Exercise for Developing and Maintaining Cardiorespiratory, Musculoskeletal, and Neuromotor Fitness in Apparently Healthy Adults: Guidance for Prescribing Exercise. Medicine & Science in Sports & Exercise 43 , 1334 (2011).19. Koriath, J. J., Lindholm, E. & Landers, D. M. Cardiac-Related Cortical Activity During Variations in Mean Heart Rate. International Journal of Psychophysiology 5 , 289–299 (1987).20. Lacey, B. C. & Lacey, J. I. Two-Way Communication Between the Heart and the Brain: Significance of Time Within the Cardiac Cycle. American Psychologist 33 , 99–113 (1978).21. Lacey, J. I. Somatic Response Patterning and Stress: Some Revisions of Activation Theory. in Pyschological Stress: Issues in Research (eds. Appley, M. H. & Trimbull, R.) (Appleton Century-Crofts, New York, 1967).22. Lacey, B. C. & Lacey, J. I. Studies of Heart Rate and Other Bodily Processes in Sensorimotor Behavior. in Cardiovascular psychophysiology: Current issues in response mechanisms, biofeedback and methodology 538–564 (AldineTransaction, New Brunswick, NJ, US, 1974).23. Lacey, J. I. & Lacey, B. C. The Relationship of Resting Autonomic Activity to Motor Impulsivity. Research Publications of the Association for Research in Nervous & Mental Disease 36 , 144–209 (1958).24. Lacey, J. I. & Lacey, B. C. Some Autonomic-Central Nervous System Interrelationships. in Physiological Correlates of Emotion (ed. Black, P.) 205–227 (Academic Press, 1970). Klimesch, W., Sauseng, P. & Hanslmayr, S. EEG Alpha Oscillations: The Inhibition–Timing Hypothesis. Brain Research Reviews 53 , 63–88 (2007).26. Romei, V., Rihs, T., Brodbeck, V. & Thut, G. Resting Electroencephalogram Alpha-Power Over Posterior Sites Indexes Baseline Visual Cortex Excitability. Neuroreport 19 , 203–208 (2008).27. Romei, V. et al. Spontaneous Fluctuations in Posterior Alpha-Band EEG Activity Reflect Variability in Excitability of Human Visual Areas. Cerebral Cortex 18 , 2010–2018 (2008).28. Jensen, O. & Mazaheri, A. Shaping Functional Architecture by Oscillatory Alpha Activity: Gating by Inhibition. Frontiers in Human Neuroscience 4 , (2010).29. Iemi, L., Chaumon, M., Crouzet, S. M. & Busch, N. A. Spontaneous Neural Oscillations Bias Perception by Modulating Baseline Excitability. J Neurosci 37 , 807–819 (2017).30. Kelly, S. P., Lalor, E. C., Reilly, R. B. & Foxe, J. J. Increases in Alpha Oscillatory Power Reflect an Active Retinotopic Mechanism for Distracter Suppression During Sustained Visuospatial Attention. Journal of Neurophysiology 95 , 3844–3851 (2006).31. Thut, G., Nietzel, A., Brandt, S. A. & Pascual-Leone, A. Alpha-Band Electroencephalographic Activity Over Occipital Cortex Indexes Visuospatial Attention Bias and Predicts Visual Target Detection. J Neurosci 26 , 9494–9502 (2006).32. Sauseng, P. et al. A Shift of Visual Spatial Attention Is Selectively Associated with Human EEG Alpha Activity. Eur. J. Neurosci. 22 , 2917–2926 (2005).33. Sandman, C. A., McCanne, T. R., Kaiser, D. N. & Diamond, B. Heart Rate and Cardiac Phase Influences on Visual Perception. J Comp Physiol Psychol 91 , 189–202 (1977).34. Jennings, J. R. & Wood, C. C. Cardiac Cycle Time Effects on Performance, Phasic Cardiac Responses, and Their Intercorrelation in Choice Reaction Time. Psychophysiology 14 , 297–307 (1977).35. Lacey, B. C. & Lacey, J. I. Cognitive Modulation of Time-Dependent Primary Bradycardia. Psychophysiology 17 , 209–221 (1980).36. Somsen, R. J. M., Jennings, J. R. & Van der Molen, M. W. The Cardiac Cycle Time Effect Revisited: Temporal Dynamics of the Central-Vagal Modulation of Heart Rate in Human Reaction Time Tasks. Psychophysiology 41 , 941–953 (2004).37. Park, H.-D., Correia, S., Ducorps, A. & Tallon-Baudry, C. Spontaneous Fluctuations in Neural Responses to Heartbeats Predict Visual Detection. Nat Neurosci 17 , 612–618 (2014).38. Cobos, M. I., Guerra, P. M., Vila, J. & Chica, A. B. Heart-Rate Modulations Reveal Attention and Consciousness Interactions. Psychophysiology 56 , e13295 (2019).39. Motyka, P. et al. Interactions Between Cardiac Activity and Conscious Somatosensory Perception. Psychophysiology 56 , e13424 (2019).40. Grund, M. et al. Respiration, Heartbeat, and Conscious Tactile Perception. J. Neurosci. 42 , 643–656 (2022).41. Vila, J. et al. Cardiac Defense: From Attention to Action. International Journal of Psychophysiology 66 , 169–182 (2007).42. Skora, L. I., Livermore, J. J. A. & Roelofs, K. The Functional Role of Cardiac Activity in Perception and Action. Neuroscience & Biobehavioral Reviews 137 , 104655 (2022).43. Corcoran, A. W., Macefield, V. G. & Hohwy, J. Be Still My Heart: Cardiac Regulation as a Mode of Uncertainty Reduction. Psychon Bull Rev 28 , 1211–1223 (2021).44. Pennebaker, J. W. & Lightner, J. M. Competition of Internal and External Information in an Exercise Setting. Journal of Personality and Social Psychology 39 , 165–174 (1980).45. Wallman-Jones, A., Perakakis, P., Tsakiris, M. & Schmidt, M. Physical Activity and Interoceptive Processing: Theoretical Considerations for Future Research. Int J Psychophysiol 166 , 38–49 (2021).46. Schandry, R. & Montoya, P. Event-Related Brain Potentials and the Processing of Cardiac Activity. Biological Psychology 42 , 75–85 (1996).47. Gray, M. A. et al. A Cortical Potential Reflecting Cardiac Function. Proceedings of the National Academy of Sciences 104 , 6818–6823 (2007).48. Park, H.-D. & Blanke, O. Heartbeat-Evoked Cortical Responses: Underlying Mechanisms, Functional Roles, and Methodological Considerations. NeuroImage 197 , 502–511 (2019).49. Coll, M.-P., Hobson, H., Bird, G. & Murphy, J. Systematic Review and Meta-Analysis of the Relationship Between the Heartbeat-Evoked Potential and Interoception. Neuroscience & Biobehavioral Reviews 122 , 190–200 (2021).50. Montoya, P., Schandry, R. & Müller, A. Heartbeat Evoked Potentials (HEP): Topography and Influence of Cardiac Awareness and Focus of Attention. Electroencephalography and Clinical Neurophysiology/Evoked Potentials Section 88 , 163–172 (1993).51. Petzschner, F. H. et al. Focus of Attention Modulates the Heartbeat Evoked Potential. Neuroimage 186 , 595–606 (2019).52. Al, E. et al. Heart–Brain Interactions Shape Somatosensory Perception and Evoked Potentials. Proceedings of the National Academy of Sciences 117 , 10575–10584 (2020).53. Wilke, M., Logothetis, N. K. & Leopold, D. A. Generalized Flash Suppression of Salient Visual Targets. Neuron 39 , 1043–1052 (2003).54. Poland, E., Bhonsle, A., Steinmann, I. & Wilke, M. Reduced Alpha Amplitudes Predict Perceptual Suppression. Sci Rep 11 , 13040 (2021).55. Brainard, D. H. The Psychophysics Toolbox. Spat Vis 10 , 433–436 (1997).56. Kleiner, M. et al. What’s New in Psychtoolbox-3. Perception 36 , 1–16 (2007).57. Oldfield, R. C. The Assessment and Analysis of Handedness: The Edinburgh Inventory. Neuropsychologia 9 , 97–113 (1971).58. Nes, B. M., Janszky, I., Wisløff, U., Støylen, A. & Karlsen, T. Age-Predicted Maximal Heart Rate in Healthy Subjects: The HUNT Fitness Study. Scandinavian Journal of Medicine & Science in Sports 23 , 697–704 (2013).59. Mehling, W. E. et al. The Multidimensional Assessment of Interoceptive Awareness (MAIA). PLOS ONE 7 , e48230 (2012).60. Clark, J. H. The Ishihara Test for Color Blindness. American Journal of Physiological Optics 5 , 269–276 (1924).61. Oostenveld, R., Fries, P., Maris, E. & Schoffelen, J.-M. FieldTrip: Open Source Software for Advanced Analysis of MEG, EEG, and Invasive Electrophysiological Data. Computational Intelligence and Neuroscience 2011 , 1–9 (2011).62. Pan, J. & Tompkins, W. J. A Real-Time QRS Detection Algorithm. IEEE Trans Biomed Eng 32 , 230–236 (1985).63. Sedghamiz, H. Matlab Implementation of Pan Tompkins ECG QRS detector. https://doi.org/10.13140/RG.2.2.14202.59841 (2014).64. Leutheuser, H. et al. Instantaneous P- and T-Wave Detection: Assessment of Three ECG Fiducial Points Detection Algorithms. in (IEEE Xplore, San Francisco, CA, USA, 2016). Vázquez-Seisdedos, C. R., Neto, J. E., Marañón Reyes, E. J., Klautau, A. & Limão de Oliveira, R. C. New Approach for T-Wave End Detection on Electrocardiogram: Performance in Noisy Conditions. BioMedical Engineering OnLine 10 , 77 (2011).66. Maris, E. & Oostenveld, R. Nonparametric Statistical Testing of EEG- and MEG-Data. J Neurosci Methods 164 , 177–190 (2007).67. Kern, M., Aertsen, A., Schulze-Bonhage, A. & Ball, T. Heart Cycle-Related Effects on Event-Related Potentials, Spectral Power Changes, and Connectivity Patterns in the Human Ecog. NeuroImage 81 , 178–190 (2013).68. Wöstmann, M., Alavash, M. & Obleser, J. Alpha Oscillations in the Human Brain Implement Distractor Suppression Independent of Target Selection. J. Neurosci. 39 , 9797–9805 (2019).69. Foxe, J. J. & Snyder, A. C. The Role of Alpha-Band Brain Oscillations as a Sensory Suppression Mechanism During Selective Attention. Front. Psychol. 2 , (2011).70. Worden, M. S., Foxe, J. J., Wang, N. & Simpson, G. V. Anticipatory Biasing of Visuospatial Attention Indexed by Retinotopically Specific Alpha-Band Electroencephalography Increases Over Occipital Cortex. J. Neurosci. 20 , RC63 (2000).71. Wagner, J., Solis-Escalante, T., Scherer, R., Neuper, C. & Müller-Putz, G. It’s How You Get There: Walking down a Virtual Alley Activates Premotor and Parietal Areas. Front. Hum. Neurosci. 8 , (2014).72. Chen, X., Cao, L. & Haendel, B. F. Human Visual Processing During Walking: Dissociable Pre- and Post-Stimulus Influences. NeuroImage 264 , 119757 (2022).73. Schandry, R., Sparrer, B. & Weitkunat, R. From the Heart to the Brain: A Study of Heartbeat Contingent Scalp Potentials. International Journal of Neuroscience 30 , 261–275 (1986).74. Pollatos, O. & Schandry, R. Accuracy of Heartbeat Perception Is Reflected in the Amplitude of the Heartbeat-Evoked Brain Potential. Psychophysiology 41 , 476–482 (2004).75. Pollatos, O., Kirsch, W. & Schandry, R. Brain Structures Involved in Interoceptive Awareness and Cardioafferent Signal Processing: A Dipole Source Localization Study. Hum Brain Mapp 26 , 54–64 (2005).76. Villena-González, M. et al. Attending to the Heart Is Associated with Posterior Alpha Band Increase and a Reduction in Sensitivity to Concurrent Visual Stimuli. Psychophysiology 54 , 1483–1497 (2017).77. Perogamvros, L. et al. Increased Heartbeat-Evoked Potential During REM Sleep in Nightmare Disorder. NeuroImage: Clinical 22 , 101701 (2019).78. Buot, A., Azzalini, D., Chaumon, M. & Tallon-Baudry, C. Does Stroke Volume Influence Heartbeat Evoked Responses? Biological Psychology 165 , 108165 (2021). Crossref Google Scholar Information & Authors Information Version history V1 Version 1 27 January 2025 Peer review timeline Published Psychophysiology Version of Record 9 Sep 2025 Published Copyright This work is licensed under a Non Exclusive No Reuse License. Collection Psychophysiology Authors Affiliations Aishwarya Bhonsle 0000-0002-7848-3845 Georg-August-Universitat Gottingen View all articles by this author Melanie Wilke [email protected] Georg-August-Universitat Gottingen View all articles by this author Metrics & Citations Metrics Article Usage 574 views 195 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Aishwarya Bhonsle, Melanie Wilke. Neural and Cardiac Contributions to Perceptual Suppression During Cycling. Authorea . 27 January 2025. DOI: https://doi.org/10.22541/au.173796146.68913752/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.173796146.68913752/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9ffa1dec28ee4193',t:'MTc3OTQzNTA0MA=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();