The temporal neurovascular coupling response remains intact during sinusoidal hypotensive and hypertensive challenges

Introduction. Neurovascular coupling (NVC) describes the coupling of neuronal metabolic demand to blood supply, which has shown to be impaired with chronic hypertension, as well as with prolonged hypotension. However, it is unknown the extent the NVC response remains intact during transient hypo- and hyper-tensive challenges. Methods. Fifteen healthy participants (9 females/6 males) completed a visual NVC task (‘Where’s Waldo?’) over two testing sessions, consisting of cyclical 30 s eyes closed and opened portions. The Waldo task was completed at rest (8 min) and concurrently during squat-stand maneuvers (SSMs; 5 min) at 0.05 Hz (10 s squat/stand) and 0.10 Hz (5 s squat-stand). SSMs induce 30–50 mmHg blood pressure oscillations, resulting in cyclical hypo- and hyper-tensive swings within the cerebrovasculature, allowing for the quantification of the NVC response during transient hypo- and hyper-tension. Outcome NVC metrics included baseline, peak, relative increase in cerebral blood velocity (CBv), and area-under-the-curve (AUC30) within the posterior and middle cerebral arteries indexed via transcranial Doppler ultrasound. Within-subject, between-task comparisons were conducted using analysis of variance with effect size calculations. Results. Differences were noted between rest and SSM conditions in both vessels for peak CBv (all p < 0.045) and the relative increase in CBv (all p < 0.049) with small-to-large effect sizes. AUC30 metrics were similar between all tasks (all p > 0.090) with negligible-to-small effect sizes. Conclusions. Despite the SSMs eliciting ∼30–50 mmHg blood pressure oscillations, similar levels of activation occurred within the neurovascular unit across all conditions. This demonstrated the signaling of the NVC response remained intact during cyclical blood pressure challenges.


Introduction
The neurovascular unit is comprised of neurons, astrocytes, pericytes, microglia, and endothelial cells, which have several functions (Schaeffer and Iadecola 2021). One of the major functions is to ensure neuronal metabolic demand receives appropriate blood supply based on energy requirements, which is commonly referred to as neurovascular coupling (NVC) or functional hyperemia (Bell et al 2020). However, in addition to the metabolic activity of the brain parenchyma, the cerebrovasculature is influenced by numerous stimuli including blood pressure changes, circulating vasoactive substances, sympathetic and parasympathetic nervous system activity, and so forth (Willie et al 2014). Typically, NVC assessments are conducted during resting paradigms (Phillips et al 2016), thus little is known regarding the extent the NVC response remains intact during systemic blood pressure challenges.
In the presence of transient elevations or reductions in blood pressure, the brain will adapt the cerebrovasculature in order to maintain cerebral perfusion, which is known as dynamic cerebral autoregulation (dCA) (Brassard et al 2021). In a review focused on the topic of cerebral blood flow regulation, Claassen et al (2021) stated if the dCA mechanisms are overwhelmed, hyperperfusion could ensue resulting in impairments to the blood brain barrier, which may lead to a reduced excitation of the NVC response. Nonetheless, work within anesthetized Felis catus, noted moderate elevations in arterial pressure (∼40 mmHg) were effectively buffered to prevent overwhelming pressure within the pial and penetrating arterioles and venules (Fog 1938). While it is known chronic hypo-and hyper-tension result in an impaired NVC response (Phillips et al 2016), it is unknown the acute impact these states have on healthy individuals.
A previous study completed by Samora et al (2020) quantified the extent the temporal NVC response is influenced by cerebral hypotension (via lower body negative pressure at −20 and −40 mmHg) and hypocapnia (via hyperventilation). Across all conditions, no differences were noted in the NVC response, unveiling the signaling of the neurovascular unit remained intact (Samora et al 2020). Nonetheless, these authors only quantified the NVC response during steady-state paradigms designed to reduce perfusion, rather than creating a hypertensive state (Samora et al 2020). Further, utilizing a steady-state paradigm allows the cerebrovascular and cardiovascular systems to adjust to the orthostatic challenge and create a new homeostatic setpoint. Conversely, large ephemeral blood pressure oscillations, such as through the use of squat-stand maneuvers (SSMs), cause the cerebrovasculature to rapidly adapt to cyclical ∼30-50 mmHg swings in blood pressure (Smirl et al 2015). Therefore, the purpose of the present study was to further the findings by Samora and colleagues (2020) and Fog (1938), by investigating the extent the NVC coupling response remains intact during transient hypo-and hypertensive challenges via SSMs (∼30-50 mmHg blood pressure oscillations). In conjunction with the previous literature, it was hypothesized the NVC response would remain intact during transient cyclical decreases and increases in blood pressure.

Ethical approval
The current study has been approved by the University of Calgary's Conjoint Health and Research Ethics Board (REB20-2112). Before study commencement, study protocol and instrumentation were explained to each participant, all questions were answered, and written-informed consent was obtained. Aside from study registration within a database, all guidelines and standards were followed in accordance with the Declaration of Helsinki and institutional guidelines.

Participants and study design
Participants were excluded if they were classified as obese, unless due to an elevated muscle mass (body mass index >35 kg m −2 ), (2) were hypertensive (systolic >145 mmHg and diastolic >90 mmHg), (3) had a history of smoking and/or substance abuse, and/or (4) were pregnant or attempting to become pregnant. As the current protocol required participants to squat with their eyes closed, participants were excluded if they were unable to successfully perform the SSMs during a 30 s pilot while participants had their eyes closed.
A total sample of 15 healthy, uninjured, university students (9 females and 6 males) were recruited to participate using a randomized crossover study design. Females were an average of 23.0 ± 4.9 years and had a body mass index 23.7 ± 5.4 kg m −2 , whereas males were an average of 25.8 ± 5.2 years and had a body mass index 24.5 ± 2.5 kg m −2 . Testing involved the completion of two ∼2.5 h sessions separated by 7.0 ± 2.2 d (range: 4-14 d). Volunteers refrained from alcohol, caffeine, vaping, smoking, cannabis, and other recreational drugs for a minimum of 12 h prior to participation (Ainslie et al 2007, Smirl et al 2015. Participants abstained from exercise for a minimum of six hours before study commencement as all aspects of cerebrovascular regulation have demonstrated to be recovered at this point following exercise (Burma et al 2020b, Burma et al 2021a, Kennedy et al 2022.

Instrumentation
Participants were fitted with a transcranial Doppler ultrasound (TCD) headframe containing two ultrasound probes (Doppler Box, DWL Technologies, San Juan Capistrano, CA, USA), which was used to measure cerebral blood velocity (CBv) within the deep conduit arteries (i.e. posterior (PCA) and middle (MCA) cerebral arteries (Willie et al 2014). The two 2 MHz ultrasound probes were placed on the transtemporal windows; where the MCA and PCA segments were insonated on the right and left sides of the head, respectively. Both vessels were located by an experienced sonographer through analyzing the depth of the signal, finger tracking, visual tasks, and use of carotid compressions (Willie et al 2014). Once confirmation the appropriate vessels were insonated took place, the probes were locked into place. Beat-to-beat blood pressure was obtained using photoplethysmography via finger cuff with height calibration, where the latter corrected blood pressure to the height of the heart (Finometer NOVA, Finapres Medical Systems, Amsterdam, Netherlands) (Omboni et al 1993, Sammons et al 2007. Electrocardiography was used to determine heart rate with a 3-lead electrocardiogram (ADInstruments, Bio Amp FE231, Colorado Spring, Colorado, USA) via lead II methodology. End-tidal partial pressures of carbon dioxide (P ET CO 2 ) were derived on a breath-by-breath basis, which provided a surrogate for arterial carbon dioxide levels. All data collected from the TCD, and physiological measures were time-synced and stored using commercially available software (LabChart Pro Version 8, ADInstruments).

Experimental protocols
Data collection occurred between November and December 2021 within the Cerebrovascular Concussion Laboratory at the University of Calgary located 1111 m above sea level. Laboratory conditions were highly controlled for: barometric pressure of 663.9 ± 6.9 mmHg; humidity at 19.5% ± 3.9%; and temperature of 21.5 ± 0.4°C. Following the equipment being attached and calibrated, participants completed a three-block randomized protocol, consisting of a complex scene search (i.e. 'Where's Waldo?') (Handford et al 1987). Participants completed the 'Where's Waldo?' paradigm a total of three times; once at rest and twice during SSMs at 0.05 and 0.10 Hz. A schematic detailing the experimental protocol for all three conditions is detailed in figure 1. To minimize the effect of task order, these tasks were completed in a randomized order, with sufficient rest/washout time between (at least 5 min) consistent with previous literature (Smirl et al 2015, Burma et al 2020a, 2020b, Burma et al 2021b. During all paradigms, participants completed several cycles of 30 s eyes closed, followed by 30 s of eyes opened task engagement where they actively searched for Waldo and the other characters in the Waldo Universe (Handford et al 1987). For each cycle during the eye-open stages, participants were given a new puzzle to ensure engagement was maintained and did not confound the measures of interest (Burma et al 2021c). For the resting condition, eight cycles were completed (∼8 min) as is common throughout the literature (Phillips et al 2016, Smirl et al 2016, Burma et al 2021c. However, due to the physiological demands of the SSMs, only five cycles of 'Where's Waldo?' were completed (∼5 min). Completing five or greater trials has been demonstrated to be highly valid when compared to eight trials (Phillips et al 2016 with similar degrees of between-day and within-day reliability . Nonetheless, the authors compared the values obtained during the resting condition between five and eight trials, finding an average coefficient of variation (CoV) of 3% for all variables. Thus, the difference in the number of trials completed between the resting condition and the two SSM frequencies (see below) had an inconsequential impact on the derived outcome measures and conclusions.
The SSMs were completed at 0.05 Hz (10 s squat/stand (20 s cycles)) and 0.10 Hz (5 s squat/stand (10 s cycles)) as transfer function analysis estimates produced at these point-estimates reveal important information surrounding the physiological influences on dCA (Panerai et al 2023). More specifically, the SSMs at 0.05 Hz are known to delineate metabolic, endothelial, and myogenic influences on cerebrovasculature (Hamner et al 2010, Tan et al 2013, Hamner and Tan 2014, whereas SSMs at 0.10 Hz reveal information regarding sympathetic innervation (i.e. Mayer wave) (Julien 2006). To maximize safety, a researcher stood near the participant as the protocol consisted of concurrent SSMs during cyclical eyes opened and eyes closed segments. No participants had any difficulties or balance issues completing the described protocol. Over the five-minutes of the concurrent SSMs and 'Where's Waldo?' tasks, participants completed 15 squats at 0.05 Hz and 30 squats at 0.10 Hz. The start of the task was randomized so for half of the participants the eyes opened portion aligned with the squat transition, while for the others it aligned with the standing transition. These differences were nonetheless controlled for, which is further detailed below.

Data processing
Peak-systolic and end-diastolic values for the PCA velocity (PCAv), MCA velocity (MCAv), and blood pressure metrics were calculated from the raw beat-to-beat traces. Given the high temporal resolution of the TCD and Finapres, mean values for the variables were calculated as the average of all data points collected within each cardiac cycle (beginning of systole to the end of diastole). Further, P ET CO 2 values were derived from the peak end-tidal carbon dioxide at the end of expiration for each breath. The R-R interval was used to determine heart rate. Artifacts within the data primarily occurred within the systolic CBv traces (<0.2%), which were filtered using the average of the values before and after the artifact. These were determined as a measured CBv systolic value inconsistent with a calculated systolic value using the mean and diastolic CBV traces.
Lastly, blood pressure, MCAv, and PCAv across the cardiac cycle (i.e. diastole, mean, and systole) were calculated at the peak and nadir of each squat and stand portion across the eyes open, and eyes closed sections. This was completed to ensure any differences in NVC results could not be explained due to differing squat depths between tasks and/or within a given task between the eyes opened/eyes closed portions of these trials. For each of the three protocols, the completed five (SSM) or eight (resting condition) trials were averaged to produce a single omnibus response for each protocol for each participant (figure 2). The outcome metrics of interest included: (1) baseline CBv averaged during eyes closed, (2) peak CBv during active task engagement, (3) percent increases in CBv from baseline to peak, and (4) the area-under-the-curve during the 30 s of task engagement (AUC30). The AUC30 metric was derived as the area enclosed by the CBv response during task activation, with the baseline CBv eyes closed portion used as the reference for the x-axis. The boundaries for the y-axes consisted of the eyes opened and eyes closed trigger for each cycle, which were averaged to produce a final 30 s response. NVC metrics were calculated within the PCA and MCA across the cardiac cycle (i.e. diastole, mean, and systole) (Burma et al 2023). The repetitive cycles were averaged where the last 20 s of the eyes closed and the entire 30 s of the eyes opened task engagement were used to quantify the aforementioned four variables. A detailed description of these parameters can be found elsewhere (Burma et al 2021c).
While previous studies have used a 20 s eyes closed and 40 s eyes opened protocol (Smirl et al 2016, Burma et al 2021a, 2021c, this was adapted in the current investigation to a 30 s eyes closed and 30 s eyes opened protocol to control for the blood pressure oscillations and enable comparable baseline metrics could be derived between all three conditions (i.e. resting, 0.05 Hz SSMs, and 0.10 Hz SSMs). For example, using a 5 s baseline would contain half of one cycle at 0.10 Hz and a quarter of one cycle at 0.05 Hz, thus the baseline measure would be confounded by blood pressure influences producing invalid comparisons. Typically, upon closing one's eyes, it takes ∼10 s for the PCAv to reach its new steady state associated with the given task (Attwell et al 2010). Therefore, with a 30 s baseline, little change in CBv would be expected to occur for the 20 s leading up to the eyes opened stimulus. This period was used to calculate baseline CBv values for all three paradigms. However, with the SSM at 0.05 Hz, it takes 20 s to complete a full cycle, meaning the eyes opened protocol would contain 1.5 cycles (figure 2). To control for this, baseline values were calculated as using the −20 to −10 s twice and the −10 to 0 s once prior to the eyes opened stimulus (figure 2) (i.e. 67% of the baseline was derived from the −20 to −10 s portion and 33% from the −10 to 0 s portion). This ensured the same aspects of the SSMs were included in both the eyes closed and eyes opened aspects (e.g. two squats (67%) and one stand (33%) or one squat (33%) and two stands (67%)), while ensuring an eyes closed steady state PCAv was achieved (figure 2). The 30 s of eyes opened, task engagement was used to determine peak CBv, the relative change in CBv, and AUC30 metrics. The outcome measures of interest were calculated using self-written Excel scripts (Microsoft, Redmond, Washington, United States).

Statistical analyzes
RStudio (version 1.4.1060) was used in the current investigation to perform all statistical analyzes. As the 15 participants completed repeat testing over two time points, data from each testing session were used independently, meaning 30 data points were available for every comparison. One-way analysis of variance (ANOVA) were used to compare differences in NVC between the three conditions (i.e. resting, 0.05 Hz SSMs, and 0.10 Hz SSMs). Post-hoc analysis through Tukey's honestly significant differences were used to determine where a difference occurred between groups, in the presence of a significant F-test. The homogeneity of variances was assessed for all comparisons using Levene's test. If this produced a significant test, data were then logtransformed and reassessed to ensure this assumption was not violated. Nonetheless, previous research has demonstrated the F-test to be highly robust in mitigating type I error (near 100%), when the included groups have equal and unequal sample sizes in each group when the data does or does not follow a normal distribution, and when the group distributions follow equal or unequal shapes (Blanca et al 2017). As there is enlarging Note: the 0.05 Hz SSMs contained one full oscillation cycle during the eyes closed baseline portion (20 s), but one and a half cycles during the eyes opened (two squats and one stand, or vice versa). Therefore, Panel E highlights how the baseline values were standardized by deriving 67% from the −20 to −10 s portion and 33% from the −10 to 0 s portion. Conversely, the 0.10 Hz SSMs contained two full oscillation cycles and three cycles during the eyes closed and eyes opened portions, respectively, and thus did not require standardization. Centimetres (cm), posterior cerebral artery velocity (PCAv), and seconds (s).
concern within physiological and biomedical-based literature surrounding the use of a binary p-value to define the presence or absence of biological processes (Panagiotakos 2008, Amrhein et al 2019, Halsey 2019, inferences were made based upon the combination of effect sizes and p-values (Bakeman 2005). For the ANOVAs, η 2 G coefficients were computed using thresholds of <0.02, 0.02-0.14, 0.14-0.26, and >0.26 to demarcate negligible, small, medium, and large effect sizes, respectively (Bakeman 2005). For the post-hoc analyzes, Cohen's d coefficients were calculated using recommendations through literature of negligible (<0.2), small (0.2-0.5), moderate (0.5-0.8), and large (>0.8) (Lakens 2013). Lastly, to understand the extent physiological variables (e.g. P ET CO 2 , heart rate, etc) differed between groups, within-subject CoV values were calculated as the quotient of the standard deviation and average of each outcome metric (Hopkins 2000). The average from all participants was used to produce the CoV point-estimate, which the 30 values were then bootstrapped with 10 000 resamples to produce the 95% confidence intervals (95% CI) for all CoV estimates.
Data were reported as mean ± standard deviation or CoV estimate with the 95% CI. Finally, despite statistical analyzes being performed on log-transformed data, the raw values are displayed within the figures to provide a clear representation of the data with respect to the values commonly reported throughout the literature. Alpha was set a priori at 0.05 for the omnibus tests and was corrected for the post-hoc comparisons based on the three tasks being compared.

Results
Physiological data Heart rate, blood pressure, and respiratory parameters from each of the tests are displayed in table 1. Greater variation was noted between tests regarding the cardiovascular metrics, as these were manipulated during the SSMs (table 1). Additionally, figure 3 demonstrates the near homogeneity of diastolic, mean, and systolic blood pressure values during both the squat and standing portion when compared between the eyes closed and eyes opened sections. This highlights any differences noted to NVC coupling metrics were not due to differences in squat depth between and within tasks (figure 3). Figure 4 highlights the change in blood pressure, MCAv, and PCAv across the cardiac cycle (i.e. diastole, mean, and systole) compared to baseline during the squat and stand portions of the SSMs. During the stand portions compared to the resting Waldo task, blood pressure values across the cardiac cycle in the standing portion were an average of −7.50 mmHg, −3.03 mmHg, and 6.5 mmHg for diastole, mean, and systole respectively. In the squat portion resulted in an average hypertensive challenge above baseline of 21.1 mmHg, 27.6 mmHg, and 41.3 mmHg for diastole, mean, and systole respectively ( figure  4). Finally, the data statistical analyzes for all tasks and both vessels are detailed in tables 2 and 3, respectively.

Discussion
The current investigation sought to delineate the extent the NVC response remains intact during rhythmically induced ∼30-50 mmHg blood pressure oscillations simulating large and robust transient hypo-and hypertensive challenges. Differences were noted to the absolute CBv metrics (i.e. baseline and/or peak) and the relative percent increase between resting and SSM conditions, being primarily attributable to the physiological CBv Figure 5. Violin and boxplots of neurovascular coupling parameters derived within the posterior cerebral artery using a 'Where's Waldo?' paradigm from 15 individuals (9 females and 6 males). These were completed at two time points producing 30 unique data points. Metrics were quantified during a resting condition (blue), as well as during squat-stand maneuvers (SSM) at 0.05 Hz (gray) and 0.10 Hz (red). * p < 0.05, ** p < 0.01, and *** p < 0.001 from the Tukey post-hoc comparisons following significant omnibus tests from analysis of variance comparisons. The main effect omnibus tests were computed across the three tasks, within each measure and cardiac cycle component. Area-under-the-curve during the 30 s of task engagement (AUC30), centimetres (cm), and seconds (s).
response associated with SSMs. Despite this, the total activation across the NVC (AUC30) was unchanged. The current results provide evidence that the NVC response was present and remained highly functional during the blood pressure perturbations, consistent with previous observations in the cerebrovascular physiology literature (Fog 1938, Claassen et al 2021.

Physiological underpinnings
SSMs are capable of causing blood pressure oscillations of ∼30-50 mmHg due to the maximal engagement of the skeletal muscle pumps, increased venous return, and an elevated ejection fraction (Smirl et al 2015). When examining the PCAv AUC30 metrics within each phase of the cardiac cycle independently, the signaling on the cerebrovasculature remained highly functional demonstrated by the null findings (i.e. total activation) ( figure 4). . Violin and boxplots of neurovascular coupling parameters derived within the middle cerebral artery using a 'Where's Waldo?' paradigm from 15 individuals (9 females and 6 males). These were completed at two time points producing 30 unique data points. Metrics were quantified during a resting condition (blue), as well as during squat-stand maneuvers (SSM) at 0.05 Hz (gray) and 0.10 Hz (red). Area-under-the-curve during the 30 s of task engagement (AUC30), centimetres (cm), and seconds (s). * p < 0.05, ** p < 0.01, and *** p < 0.001 from the Tukey post-hoc comparisons following significant omnibus tests from analysis of variance comparisons. The main effect omnibus tests were computed across the three tasks, within each measure and cardiac cycle component. Area-under-the-curve during the 30 s of task engagement (AUC30), centimetres (cm), and seconds (s).
Moreover, figure 4 highlights the systolic elevations were ∼41 mmHg compared to baseline, which aligns with the notions stated by Claassen et al (2021) where NVC still occurred, albeit to a slightly lesser extent. Here the NVC response was still present but began to slightly diminish at this ∼40 mmHg hypertensive threshold. Future research is warranted to delineate if there is a threshold that results in the NVC response via AUC30 becoming overwhelmed/abolished, as well as further investigations into the precise physiological underpinnings acute and chronic hypo-and hyper-tensive have on the NVC response.
These findings are homogenous to Samora et al (2020) who demonstrated the NVC response remained highly functional during steady-state blood pressure challenges via lower body negative pressure (−20 and −40 mmHg). An important distinction between studies is the nature of the blood pressure changes as SSMs produce transient hyper-and hypo-tensive states; whereas lower body negative pressure strictly produces a mild cerebral hypotensive challenge that the brain must buffer against. Despite the methodological differences similar findings were produced, denoting it appears the coupling of neuronal activation to blood flow demand remains highly linked despite both steady-state and ephemeral hypotensive/hypertensive challenges.
The increased peak PCA (figure 4) and MCA (figure 5) velocity metrics across all phases of the cardiac cycle were attributed to the ephemeral increase in CBv associated with the nature of the SSMs (Birch et al 1995, Claassen et al 2009, Smirl et al 2015. As the percent increase is derived based on the baseline and peak values, the differences in this metric across the cardiac cycle would be attributed to the SSMs. However, the only differences noted in baseline metrics were within the PCA, when the cerebrovasculature was pressure-passive (i.e. diastole) ( figure 5(A)). This finding was likely due to a combination of the anatomical differences between vessels insonated and the interplay between tasks (Zarrinkoob et al 2015). Previous work using magnetic resonance imaging has highlighted that ∼70% of CBF arises from the carotid circulation (i.e. MCA), whereas ∼30% comes from the vertebral artery (Zarrinkoob et al 2015). When quantified with TCD, the PCAv is lower than the MCAv, resulting in a physiological phenomenon known as critical closing pressure (CrCP) occurring more frequently within the PCA than the MCA (Panerai 2003), characterized as CBv falling to 0 cm s −1 during a TCD recording. The CrCP describes the intrinsic desire of a vessel to collapse upon itself when pressure is not sufficient (Panerai 2003). During the standing portions of the SSMs, the skeletal muscles relax, causing blood to pool in the venous system of the extremities (Smirl et al 2015). This leads to a reduced CBv, which would increase the occurrence of CrCP, especially within a vessel with lower blood flow and/or pressure (i.e. PCAv during diastole) (Panerai 2003). Moreover, this would have been further exacerbated by the eyes closed portion of the NVC task concurrently reducing CBF to a greater extent within the PCA, as this vessel would have responded to the diminished metabolic demand of the visual cortices (Smirl et al 2016). Taken together, there is a strong physiological basis for the occurrence of a difference in PCAv diastolic baseline resulting from a combination of venous pooling and reduced metabolic demand associated with one having their eyes closed ( figure 5(A)). Finally, the presence of CrCP resulted in slightly lower baseline diastolic and mean CBv values, explaining the relative percent increases greater than 100% in some participants.
Clinical relevance and implications for future research While a debate persists surrounding the best approach to quantify dCA (Simpson andClaassen 2018, Tzeng andPanerai 2018), the current study also provides further evidence for the use of driven measures, and specifically SSMs, for dCA assessments. It has been previously suggested induced blood pressure oscillations may produce autonomic, respiratory, and or other physiological alterations that may have an impact on myogenic and sympathetic regulation of the endothelium (Tzeng and Panerai 2018). However, the current investigation highlighted the cyclical blood pressure oscillations did not impact the signaling of the neurovascular unit, lessening the concern induced oscillations overwhelm the autoregulation mechanism of the cerebrovasculature. A further consideration surrounds aging, and though the literature has noted minimal age-related differences exist within the pressure-flow relationship, utilizing a dual-task paradigm (i.e. 'Where's Waldo?' and SSMs) may be influenced by physiological adaptations associated with aging (e.g. arterial stiffening, reduced baroreceptor sensitivity, lowered CBv, etc). Therefore, further work is a necessity to see if these results are similar to elderly and/or clinical populations.

Limitations
A limitation surrounding the neuroimaging technique and methodological approach utilized is that TCD is unable to quantify cross-sectional area (Ainslie and Hoiland 2014). As SSMs cause ∼30-50 mmHg blood pressure perturbations (Smirl et al 2015), it is likely the diameter of the vessel may have changed during the Waldo SSM tasks compared to Waldo at rest, thus reducing the ability to make definite conclusions. Nevertheless, regardless of this concern, no differences in AUC30 metrics were present highlighting despite a potential change in the vessel cross-sectional area, the NVC response was maintained. A benefit of employing TCD in the current investigation is its high temporal resolution and ability to collect data during dynamic movements (Purkayastha and Sorond 2012). This enabled for the a priori question to be answered using the most robust methodological approaches. It should be noted neuronal activation was not directly measured, rather a surrogate was used via changes in PCAv from rest to task engagement. However, extensive preliminary work has been completed to ensure the 'Where's Waldo?' task paradigm increases CBv to a greater extent within the PCAv (i.e. visual processing networks) compared to the MCAv (i.e. motor-based functions, goal-oriented behavior, etc). As well, end-tidal carbon dioxide levels were monitored on a breath-by-breath basis to ensure participants were within eucapnia across all conditions (table 1). Hormones, fitness, and numerous other covariates were not directly controlled; however, the comparisons were performed within each subject on the same day via a randomized crossover design (Mills et al 2009). This reduces the likelihood these would have impacted the outcome variables. Future studies should seek to control these factors, especially when making between group interpretations. Further, the SSM tasks were completed in an upright position, while the resting Waldo was seated. The cerebellum (i.e. the region associated with balance and postural control) (Takakusaki 2017) may have also been activated to a greater extent during the SSMs, which is fed through the posterior circulation (Chandra et al 2017). Albeit this likely had an inconsequential impact compared to the metabolic demands associated with the Waldo task. Finally, it should be noted repeat measures from the same participants were analyzed as independent values comparing the three Waldo tasks. This artificially inflates the power of the current investigation; however, the findings were homogeneous with previous literature with rationale physiological explanations for the conclusions, suggesting these likely would hold with a large sample size.

Conclusions
The present investigation sought to empirically assess the notion that SSMs may override the autoregulatory and NVC capabilities of the brain. A visual NVC task was performed concurrently during these maneuvers, where total activation measured via the AUC30 metric was similar between all protocols. This confirms the notion that the neuronal signaling of the neurovascular unit remained responsive despite the systemic influences of large rhythmic oscillations in blood pressure. Nonetheless, it did appear slight cardiac cycle differences were present, with systole showing slightly less activation compared to diastole or mean; however, further work is required to tease apart if an acute hypertensive threshold exists where the NVC response is unable to function appropriately. In conclusion, the NVC response remained highly functional in healthy young individuals during ephemeral hypotensive and hypertensive challenges.
the Natural Sciences and Engineering Research Council (Alexander Graham Bell Canada Graduate Scholarship-Doctoral Program).