Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

The scalability of common paradigms for assessment of cognitive function: A functional transcranial Doppler study

  • Kannakorn Intharakham ,

    Roles Data curation, Formal analysis, Investigation, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Department of Cardiovascular Sciences, University of Leicester, Leicester, United Kingdom

  • Ronney B. Panerai,

    Roles Conceptualization, Formal analysis, Methodology, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliations Department of Cardiovascular Sciences, University of Leicester, Leicester, United Kingdom, NIHR Leicester Biomedical Research Centre, British Heart Foundation Cardiovascular Research Centre, Glenfield Hospital, Leicester, United Kingdom

  • Thompson G. Robinson

    Roles Conceptualization, Methodology, Resources, Supervision, Writing – review & editing

    Affiliations Department of Cardiovascular Sciences, University of Leicester, Leicester, United Kingdom, NIHR Leicester Biomedical Research Centre, British Heart Foundation Cardiovascular Research Centre, Glenfield Hospital, Leicester, United Kingdom


Cognitive paradigms induce changes in cerebral blood flow (CBF) associated with increased metabolic demand, namely neurovascular coupling (NVC). We tested the hypothesis that the effect of complexity and duration of cognitive paradigms will either enhance or inhibit the NVC response. Bilateral CBF velocity (CBFV) in the middle cerebral arteries (MCAs) via transcranial Doppler ultrasound (TCD), blood pressure (BP), electrocardiogram (ECG) and end-tidal CO2 (EtCO2) of 16 healthy participants (aged 21–71 years) were simultaneously recorded at rest and during randomized paradigms of different complexities (naming words beginning with P-,R-,V- words and serial subtractions of 100–2,100–7,1000–17), and durations (5s, 30s and 60s). CBFV responses were population mean normalized from a 30-s baseline period prior to task initiation. A significant increase in bilateral CBFV response was observed at the start of all paradigms and provided a similar pattern in most responses, irrespective of complexity or duration. Although significant inter-hemispherical differences were found during performance of R-word and all serial subtraction paradigms, no lateralisation was observed in more complex naming word tasks. Also, the effect of duration was manifested at late stages of 100–7, but not for other paradigms. CBFV responses could not distinguish different levels of complexity or duration with a single presentation of the cognitive paradigm. Further studies of the ordinal scalability of the NVC response are needed with more advanced modelling techniques, or different types of neural stimulation.


Neurovascular coupling (NVC) describes the mechanism of cerebral blood flow (CBF) regulation through changes in arteriolar and capillary diameter to match increasing metabolic demand during neural activation [1, 2]. Action potentials, resulting from neural stimulation, release neurotransmitters to the neurovascular unit (neurons, astrocytes, pericytes, endothelial cells and extracellular matrix components), where neuronal mediators, such as K+, NO and adenosine, control vascular smooth muscle contraction or relaxation [35]. CBF changes in response to cognitive task activations can be studied with a number of non-invasive methods, including magnetic resonance imaging (MRI) [6], positron emission tomography (PET) [7], single-photon emission computed tomography (SPECT) [8], near infrared spectroscopy (NIRS) [9], laser Doppler flowmetry [10] and transcranial Doppler ultrasound (TCD) [11]. MRI, PET and SPECT may cause patient discomfort, and are not suitable for screening and follow-up studies. However, TCD can measure CBF velocity (CBFV) at rest, or during cognitive tasks, usually referred to as functional TCD (fTCD) [12]. fTCD has multiple advantages in terms of low-cost, high temporal resolution, portability to the bedside and non-harmful repetitive studies. Therefore, current TCD applications have been adopted to study a wide range of conditions such as sickle cell disease [13], ischaemic [14] or haemorrhagic [15] stroke, brain stem death [16], intraoperative monitoring [17], cerebral microembolism in right-to-left cardiac shunt [18], and cognitive impairment [19], including cerebral pressure autoregulation [20] and neuroactivation [21]. The quantification of the NVC response has been assessed by CBFV changes (ΔCBFV), induced by neural stimulation [12, 2225]. Moreover, fTCD has been extensively used for assessing NVC in the lateralization of functional brain regions with a range of tasks, such as cognitive [23, 2628], visual [11, 29, 30] and sensorimotor [24, 31, 32] paradigms.

In addition to detecting differences in lateralisation, a diagnostic tool to assess cognitive function needs to satisfy the condition of scalability, that is, to demonstrate the sensitivity to reflect differences in cognitive load, or its complexity. In the case of fTCD, this has been shown to be the case with visual stimulation [30, 33], with changes in ΔCBFV detected in the posterior cerebral artery (PCA). With cognitive paradigms though, evidence of scalability is limited, and far from generalizable. In a few studies, fTCD has shown sensitivity to detect differences in difficulty and cognitive load, with the use of n-back paradigms [28, 34, 35]. The n-back paradigm stimulates specific aspects of cognition, involving attention and short-memory [36], but does not involve other aspects of cognitive function, such as language, fluency and visuospatial domains, which are part of a more comprehensive assessment of cognitive function [3739]. Previous studies have shown that fTCD can detect the NVC response to all five main cognitive domains of the Addenbrooke’s Cognitive Examination (ACE-III) battery of tests, an important step to demonstrate the way fTCD might contribute to improving the diagnostic accuracy of routine cognitive assessment [26, 40, 41]. Given that the ACE-III battery involves 20 different tasks, assessing the scalability of each of these paradigms is beyond the capacity of a single study. As a first step, we have chosen to study the scalability of naming words (NW) and serial subtraction (SS) as these tasks have been shown to induce consistent CBFV responses [26, 41] and good reproducibility [40]. The relevance of NW and SS is that they are part, not only of the ACE-III, but also of other current methods of cognitive assessment such as the Mini Mental State Examination (MMSE) and Montreal Cognitive Assessment (MoCA) [3739].

In summary, we tested the hypotheses that the complexity and duration of stimulation of NW and SS tasks can: 1) induce either larger or smaller magnitude of NVC response, and 2) lead to inter-hemispherical differences that could be the consequence of either enhanced or inhibitory mechanisms.

Materials and methods

Subjects and measurements

Sixteen healthy participants (eight women) without history of cardiovascular, neurological, or respiratory conditions, including metabolic abnormalities, inflammatory abnormalities, and severe somatic (e.g., cancer) or psychiatric (e.g., psychotic or bipolar), were recruited from the University of Leicester as students or staff. The Research Ethics committee approved the study (ref: 19452-rp9-ls:cardiovascularsciences), and fully informed, written consent was signed by each participant. Only adults 18 years old or older, with good insonation of bilateral temporal windows were suitable for study inclusion. The Edinburgh inventory was used for assessment of handedness [42].

All participants abstained from caffeine, alcohol, vigorous exercise, heavy meal, and smoking for at least 4 hrs before attending a temperature (22–24°C) and light controlled laboratory. All measurements were undertaken with participants in the seated position. To avoid any artefacts due to movement, participants were asked to sit comfortably and avoid any movement, and all signals were checked prior to recording. The 2MHz TCD probes (Viasys Companion III device) were securely placed over the temporal windows using a head frame, and bilateral CBFV was monitored continuously in the middle cerebral artery (MCA) with the depth of insonation at 45–55 mm. Arterial volume clamping of the left middle finger digital artery was used for continuous beat-to-beat BP measurement (Finometer, Finapres Medical Systems; Amsterdam, the Netherlands), and automated brachial BP was also intermittently recorded in the right arm to calibrate the Finometer before each recording, using a validated OMRON (UA767) device. Continuous measurements were recorded for heart rate (HR), with a three-lead electrocardiography (ECG), respiratory rate and end-tidal CO2 (EtCO2), using small nasal cannula (Salter Labs, ref 4000) with a capnograph (Capnocheck Plus).

Paradigms of different complexity and duration

Prior to the tests, the paradigms were described in detail, and participants were encouraged to provide their best mental effort when providing their answers in a low voice. Each paradigm was recorded during 60s with subjects asked to keep an ‘empty mind’, Tasks were given verbally to participants during recording before task initiation. Thereafter, paradigms were performed over one of three different periods of duration stimulation, with a 60s recovery phase. Naming words (e.g. P-words) and serial number subtraction (subtract sequentially 7 from 100) paradigms have been applied routinely for diagnosing mild cognitive impairment (MCI) and dementia, and have been increasingly adopted as equivalent cognitive paradigms for assessment of NVC and also functional cerebral lateralization [26, 41]. Hence, both types of paradigms were modified to determine the effect of complexity and duration in this study. For modulating intensity of neuronal activations, naming words (NW) beginning with P-, R- and V- were adopted to represent increasing levels of difficulty corresponding to easy, moderate and hard, respectively [43]. Also, three levels of difficulty in sequential subtractions (SS) were created with SS of (100–2), (100–7) and (1000–17), respectively (Fig 1).

Fig 1. Protocol and experimental setup.

Changes in cerebral haemodynamics were recording with transcranial Doppler (TCD) insonated bilateral middle cerebral arteries (MCAs) during cognitive stimulation. Blood pressure (BP), Heart rate (HR) and End-tidal CO2 (EtCO2) were simultaneously recorded.

Subjects rested for 15 min and a resting baseline recording was performed for 5 min (Fig 1). Two separate continuous recordings of cognitive paradigms were then obtained for duration and complexity data. R-words and subtraction (100–7) were used as a reference for representing word-naming (NWREF) and serial subtraction (SSREF), respectively. Paradigms of different complexity were randomized and presented during 60s. For changing the duration of presentation, the NWREF and SSREF paradigms were also randomised and then interrupted after 5, 30 or 60s, without participants’ prior knowledge. The 60 s has been the traditional duration of NW and SS tasks [26, 41], but a plateau is usually observed from 30 s onwards and using 5 s would only detect the potential effects of central command, before the metabolic deficit starts to act as a feedback error to correct the need for additional O2 supply. The beginning and end of stimulation were marked with an on/off electrical switch to identify the duration of the paradigm and allow synchronization for coherent averaging of CBFV responses. All physiological signals were sampled at 500 Hz and recorded in the PHYSIDAS data acquisition system (Department of Medical Physics, University Hospitals of Leicester NHS Trust, Leicester, UK) for offline analysis.

Data analysis

BP was calibrated at the beginning of each recording, and all signals were visually inspected to identify artefacts, with narrow (<0.1s) spikes, which were linearly interpolated. Bilateral CBFV signals were passed through a median filter with a window width of five samples to remove random noise. All signals were low-pass filtered by a zero-phase Butterworth filter with a cut-off frequency of 20 Hz. RR interval was automatically marked from the ECG, and HR was continuously plotted against time and manually corrected under visual inspection in case of missed markings. Mean values of CBFV and BP were calculated for each cardiac cycle. Linear interpolation was applied to obtain estimates of EtCO2 synchronized to the end of each cardiac cycle. Beat-to-beat data were spline-interpolated with a third order polynomial and resampled at 5 Hz to obtain signals with a uniform time base.

Each variable was synchronized by the start of each paradigm, to obtain the coherent average of CBFV for each subject. Bilateral CBFV were expressed in percent (%) by normalization by the 30s of baseline before the paradigm activation. The synchronization placed the beginning of each paradigm at 60s, within a window of 180s duration. Only bilateral CBFV signals were averaged for 10s during five time segments, with the beginning of stimulation starting at 60s: baseline (T1) at 40-50s, (T2) at 60-70s, (T3) at 110-120s, (T4) at 120-130s, and (T5) at 160-170s. Therefore, T1 represents baseline values before stimulation, T2 the immediate response to stimulation, T3 the sustained response, T4 near the end of the response (for 60 s duration of stimulation) and T5 during recovery. All timepoints were fixed for all paradigms, independently of the duration of presentation.

Statistical analysis

All data were tested for normality using Shapiro-Wilk test. Continuous data with a Gaussian distribution are presented as mean (SD). Outliers of parameters were determined by using the boxplot with interquartile limits between 25th and 75th percentile (interquartile range, IQR) for indicating extreme values beyond 1.5 times (above or below) of the IQR as outliers which were modified by replacing the second smallest or largest values in observations to the threshold boundaries as recommended in Winsorization [44]. Paired t-test (parametric data) or Wilcoxon test (nonparametric data) were used to detect differences in hemispherical comparisons of CBFV response to each paradigm. Two-way ANOVA was adopted to test for differences between time segments and paradigms for interaction in each bilateral hemisphere. Post hoc analysis with Tukey‘s was applied for multiple comparisons. Analyses were conducted using SPSS version 25 for Windows or Graphpad Prism version 9.2 for Windows. Statistical significance was adopted at a value of p < 0.05.


Sixteen healthy participants (eight women) were recruited (mean age 31.6 years, [range 21–71 years] and mean BMI 22.6 ± 3.0 kg/cm2), and completed all the NW and SS paradigms of variable duration and complexity. A majority of participants (n = 14) were right-handed and two subjects were ambidextrous, following assessment with the Edinburgh handedness inventory. Half of all participants had Asian background, six were Caucasian, and two subjects had Caribbean and African ancestry, respectively. Eight people abstained from alcohol and two subjects were current smokers. Absolute mean (SD) values of baseline CBFV did not show a significant difference between right hemisphere (RH) and left hemisphere (LH) at 66.8 (11.4) and 64.8 (11.4) cm/s, respectively. Baseline values for mean BP, HR and EtCO2 were 89.7 (9.4) mmHg, 70.8 (8.1) bpm and 38.5 (3.1) mmHg, respectively.

Temporal CBFV responses during cognitive paradigms

The overall paradigm-synchronized patterns of bilateral cerebral haemodynamic coherent averages for the whole population with increasing complexity of NW (Fig 2A–2C) and SS (Fig 2D–2F) showed considerable similarity. A rapid increase in bilateral CBFV was observed at the start of stimulation (grey bar), with the largest peak around T2 and reducing rapidly towards baseline values before T3, around 25s after stimulation. There was a further small rise in CBFV around T3, except for SS (100–2), and a steady return to baseline after the paradigm was completed at approximately 20s. The 2-way ANOVA confirmed a highly significant effect of time for NW and SS (Fig 2 and Table 1, p < 0.001).

Fig 2.

Average population responses (n = 16) in the right (continuous line) and left (dashed line) hemispheres for the complexity of NW (A, B, C) and SS (D, E, F) paradigms of CBFV, respectively. Error bar represents the largest ±SD at the point of occurrence. Changes are shown before, during (grey bar) and after 60s of P-word (A), R-word (B), V-word (C), 100–2 (D), 100–7 (E) and 1000–17 (F). Responses were synchronized by the beginning of stimulation, set at 60s. For statistical analyses, shorter segments (T1-T5) were averaged from 40-50s (T1), 60-70s (T2), 110-120s (T3), 120-130s (T4) and 160-170s (T5), respectively.

Table 1. Mean (SD) values of bilateral cerebral hemodynamic changes at each timepoint for increasing complexity of NW and SS paradigms.

Changes in paradigm duration, also led to similar patterns of the CBFV bilateral coherent averages, for both the NWREF (R-word) (Fig 3A–3C) and SSREF (100–7) tasks (Fig 3D–3F). A steep increase to the highest peak at T2 for all durations was observed bilaterally. Thereafter, there was a reduction to baseline, though a rebound increase in bilateral CBFV to a second peak at T3 was seen for the 60s duration paradigm. The 2-way ANOVA corroborated the change in bilateral CBFV response to different duration stimulations, with a highly significant effect of time for NWREF and SSREF (Table 2, p < 0.001).

Fig 3.

Average population responses (n = 16) in the right (continuous line) and left (dashed line) hemispheres for the duration of NWREF; R-word (A, B, C) and SSREF; 100–7 (D, E, F) paradigms of CBFV, respectively. Error bar represents the largest ±SD at the point of occurrence. Changes are shown before, during (grey bar) and after each duration of 05s of NWREF (A), 30s of NWREF (B), 60s of NWREF (C), 05s of SSREF (D), 30s of SSREF (E) and 60s of SSREF (F). Responses were synchronized by the beginning of stimulation, set at 60s. For statistical analyses, shorter segments (T1-T4) were averaged from 40-50s (T1), 60-70s (T2), 110-120s (T3) and 120-130s (T4), respectively.

Table 2. Mean (SD) values of bilateral cerebral hemodynamic changes at each timepoint for different duration of NWREF and SSREF paradigms.

Effect of task complexity on cerebral haemodynamic responses

Increases in the complexity of NW and SS tasks induced similar temporal CBFV changes during neural activation as no obvious difference was seen in its effect on CBFV responses within each time period of stimulation. Table 1 presents values of bilateral CBFV changes at each time segment for both NW and SS tasks, in both RH and LH. There was no significant interaction effect on bilateral CBFV responses between complexity paradigms and time segments on two-way ANOVA (Table 1). There were no significant differences between CBFV changes for NW and SS complexities at each time-point.

Effect of task duration on cerebral haemodynamic responses

Differences in the duration of NWREF and SSREF tasks induced a similar pattern of temporal CBFV responses (Fig 3), with a lack of significance in its effect on CBFV changes at initial NVC activation (Table 2). Both RH and LH showed highly significant bilateral CBFV changes for both NWREF and SSREF (Table 2). A significant interaction was seen between timepoint and duration for the SSREF in both RH (p = 0.002) and LH (p = 0.0057). As expected, due to the reduced duration of paradigms, significant differences were seen for the SSREF with increasing durations in the RH (p = 0.02) at T3 and T4 and borderline significant in the LH (p = 0.059) (Table 2), but not for NWREF.

Interhemispheric differences

Overall, there were contradictory responses between RH and LH regarding the dominance of CBFV responses during SS and NW tasks, respectively. Significant interhemispheric differences were observed for RH with SS and LH for NW tasks but only for the timepoints T3-T5 (Table 1). In paradigms of different complexity, CBFV response for LH was significantly greater than RH (p = 0.028) during R-words at T3. On the other hand, for SS, RH had significantly higher CBFV response than LH at T4 (100–2, p = 0.041 and 100–7, p = 0.043) and T5 (1000–17, p = 0.007). However, hemispherical differences in both NWREF and SSREF were not seen at each time segment during 05s and 30s durations.


Main findings

As a non-invasive tool, fTCD has often been used to detect changes in cerebral haemodynamic responses during neural activation using cognitive paradigms [11, 21, 23]. In this novel study of the effects of paradigm complexity and duration, we focused on the coherent averages of CBFV, as the dominant technique that has been used for assessment of NVC with fTCD. Our main finding was that in healthy participants, the temporal CBFV response failed to demonstrate sensitivity to NW and SS cognitive paradigms of varying complexity and duration. By showing similar patterns of otherwise highly significant alterations in the CBFV response at the onset of all tasks, these changes, and the subsequent temporal evolution of the CBFV response was not consistently sensitive to detect any effect of complexity or duration with both the NW and SS paradigms. However, different CBFV responses were observed during the sustained response and at the end of stimulation, depending on the type of paradigm undertaken. More work is therefore needed to obtain a better understanding of the NVC response to these commonly used cognitive tasks and the poor sensitivity they express to changes in cognitive load.

Temporal pattern characteristics of fTCD responses to cognitive tasks

The highly significant bilateral CBFV differences observed across time (Tables 1 & 2), confirmed that these paradigms could induce NVC responses to cognitive stimulation for all subjects. The two distinct phases of the bilateral CBFV responses to paradigms of different complexity that we observed, are in agreement with the existing literature [45, 46]. Different interpretations have been given as to the mechanisms underlying the fast response observed at T2. Whilst the CBFV changes may represent effort-related cognitive processes [47], Boban et al. (2014) and Szirmai et al. (2005) proposed that fast-neurogenic regulation, with a more attention-induced arousal reaction to mental activations, resulted in the rise of perfusion pressure rather than representing a vasodilatory response to meet metabolic demand [48, 49]. However, Sato et al. (2001) suggested that the initial fast response at T2 resulted from direct neurogenic vasodilation resulting from free circulating acetylcholine, activation of both muscarinic and nicotinic cholinergic receptors in the cortex, and involvement of NO release from nitric oxide synthase cells such as interneurons or endothelial cells [50], which is supported by the existence of corticothalamic axons in the vascular bed of the thalamus [51]. Neurogenic regulation may therefore nourish a rapid adaptive response with uncoupling of the cortical metabolic rate during initial stimulation.

For the second peak, a gradual CBFV increase was seen as the sustained response (T3), with differences related to the complexity of NW and SS, except for SS (100–2). This slow phase of the CBFV response could reflect metabolic regulation that induced decreases in local microvascular resistance depending on the actual level of tissue CO2 concentration [49]. However, in SS (100–2), the absence of a second peak might be due to this being the easiest paradigm [52], combined with a desire to quickly complete the task with rapid answers and associated hyperventilation. Therefore, an attenuated CBFV response might be particularly associated with retrieving arithmetic facts in simple calculations

Influence of paradigm complexity on fTCD responses

Paradigms of different complexity induced CBFV responses with similar temporal patterns, and without significant differences in the amplitude of the CBFV changes, or evidence of lateralization (Table 1). Although our cognitive-challenging tasks did not provide CBFV responses with consistent interhemispheric differences, our findings were in line with other studies showing reduced lateralisation with increasing task difficulty, possibly due to recruitment of additional neural networks compensating for increased cognitive demands [53]. Also, most paradigms used in common cognitive assessments have not resulted in significant lateralisation [26]. On the other hand, there is evidence in the imaging literature that other techniques for NVC assessment might be more responsive to differences in cognitive effort. In fMRI investigations, the consequences of increasing complexity in verbal fluency performance showed greater activation in the anterior cingulate region [54], with activation predominantly in the right cerebellar hemisphere [55], and with laterality increased in the posterior part of the inferior frontal gyrus during mentally complex calculation [56]. This might help to explain our results with increasing complexity of NW and SS tasks, respectively. However, only the inferior frontal gyrus is part of the territory perfused by the MCA [57]. Thus, changes in CBF, resulting from paradigms of different complexity, might occur in regions other than those perfused by the MCA. As reported by Smirl et al. (2016), using different levels of challenge in visual search paradigms, it was possible to discriminate ΔCBFV responses in the posterior cerebral artery (PCA), but ΔCBFV changes to visual paradigms of different complexity could not be detected in the MCA [58]. Similarly, bilateral CBFV changes showed a larger response during the Stroop test with cognitive incongruent (difficult) than congruent (easy) stimuli [48]. With more demanding tasks, larger ΔCBFV was seen bilaterally during n-back paradigms compared to less demanding tasks [35]. Also, the differences between interhemispheric activations [49, 59, 60] could be discriminated by task difficulty. These pronounced differences probably depend on duration of stimulation, type and complexity of paradigms [26].

Noteworthy, the majority of previous studies used repeated presentations of the same paradigm to improve signal-to-noise ratio (SNR) of intra-subject averages of the ΔCBFV response. As seen in previous studies, presentation of complex tasks was repeated a variable number of times, including < 5 [59], 5 to 10 [24, 48, 49, 5456, 61, 62] or > 10 times [60]. Studies with repeated task presentation were likely to provide more consistent responses across individuals, as compared to the present study where tasks were presented only once.

There are potential additional explanations for the lack of significance in our results to the effects of cognitive load with NW and SS paradigms. First, the differences in complexity of paradigms in the current study might not have been sufficient to induce differential CBFV responses, as proposed in terms of the compensation-related utilization of neural circuits hypothesis (CRUNCH) to recruit more neural activations matching with load increases [63, 64]. As reported for other neuroimaging techniques (fMRI or SPECT), it provided greater activation in some specific brain areas, with increased task difficulty with verbal [54], arithmetic [56] or visual [30, 65] paradigms. Although NVC assessment with fTCD has been rated as equivalent to that obtained with fMRI [66], this does not seem to be the case in our study, with respect to the ability of fTCD to reflect differences in complexity. We can speculate that increasing tasks difficulty cannot induce the CRUNCH effect on TCD-measured CBFV changes.

Secondly, the phenomenon of ‘central command’ anticipation of the brain when preparing to receive an increased demand for oxygen may be important. To our knowledge, the role of the ‘central command’ during cognitive paradigms in fTCD studies has received very little attention in the human NVC literature [67]. Previous studies suggested that the preparatory process of blood flow adjustment involves alertness of attentional function [68], correlated with increasing CBF response before the onset of exercise that is likely governed by the ‘central command’ network [69]. Therefore, increased CBFV at the onset of all paradigms, independently of complexity or duration, was probably determined by the effects of ‘central command’ on the neurovascular unit.

Thirdly, cognitive stimulation of some brain areas leads to inhibition of other cortical centres [70]. Despite the cerebral cortex circulation supporting the local metabolism during cognitive activation [71], there is evidence in the literature showing a decreased activation in the medial prefrontal, precuneus, posterior cingulate and left angular gyrus, near to Wernicke’s area (language function), following increased cognitive load of NW tasks [72]. Similarly, Menon et al. (2010) highlighted a shift in the process of brain response from prefrontal cortex to the bilateral posterior parietal cortices, including additional brain areas (caudate nucleus and cerebellum) during complex mental arithmetic [73]. It is plausible that the impact of increasing cognitive complexity might be a true decrease in some local brain areas [72, 73]. The expected increased response in CBFV following stimulation might therefore be overshadowed by the increase in inhibition with greater cognitive demand. Finally, increases in the amplitude of the CBFV response, following cognitive paradigms of increasing complexity, could have been confounded by concomitant changes in systemic covariates. The influence of sympathetic activation inducing a rise of BP and HR during cognitive paradigms has previously been reported, and changes in EtCO2 with hypocapnia-induced hyperventilation or hypercapnia-induced breath holding can also accompany CBFV alterations during cognitive tasks [21, 74]. Furthermore, increases in HR were reported at the beginning of more complex cognitive tasks [48], resulting from arousal or stress. The contribution of BP and PaCO2 to CBFV changes following cognitive tasks has been confirmed in previous fTCD studies [75]. However, whether these co-variates can obscure the changes in CBFV response that could be expected with increasing complexity needs to be investigated with the use of more advanced multivariate dynamic modelling techniques.

Influence of paradigm duration on fTCD responses

For our hypothesis regarding paradigms of different durations, it would be expected that shorter or extended duration paradigms would result in different magnitude changes of CBFV following cognitive activation. With different durations for both NWREF and SSREF, temporal CBFV responses showed similar patterns at the first peak of initial stimulation (T2), without significant differences in the amplitude of CBFV or lateralized dominance. Only CBFV responses to the longer (60s) stimulations led to a second peak for SSREF (Table 2). Despite the 60s stimulation being more likely to induce greater metabolic demand, than 5s and 30s durations, the magnitude of CBFV changes was lower than for stimulation with 30s duration (Fig 3). These results were consistent with previous studies of different durations of stimulation, which demonstrated that the longer duration stimulation would not necessarily maximize the CBFV magnitude change [29, 76].

To our knowledge, there have been no previous fTCD studies of the effects of changing cognitive paradigm duration on the induced changes in CBFV. However there have been previous TCD studies using visual stimulation of different durations [62, 76, 77]. The results of these studies were consistent, showing an identical curve of CBFV response at the initial phase, independent of the duration of stimulation, as seen in other imaging literature (e.g. BOLD) [10, 78, 79] and our paradigms. According to the ‘central command’ hypothesis, the lack of marked differences in CBFV responses with varying durations could be explained by the anticipatory response to cognitive demand in the attentional process [68], governing an upsurge of CBFV around 5s after onset stimulation, without further influence of the duration of stimulation on the NVC mechanism.

In tasks with 60s duration, a slow CBFV peak was found in both hemispheres before T3 in SSREF (Fig 3). This is in contrast to CBFV patterns seen in previous studies with stimuli of varying duration, that demonstrated a plateau phase with longer duration stimulation, using both fTCD [62] and BOLD [79]. The difference in CBFV responses might depend on the type of paradigm. The plateau curve might relate to an adaptative brain response as a compensatory function [80, 81], but the two-peaked pattern in CBFV response seen in our and other studies [46, 48] might be confounded from involuntary hyperventilation-induced hypocapnia after the first peak [45, 48, 82] which affects approximately 30s of the magnitude of the CBFV response. Accordingly, a second peak will be missing in paradigms of shorter (5s and 30s) duration.

Clinical implications and limitations

This is the first systematic study to determine fTCD-measured NVC responses to NW and SS paradigms of varying complexity (easy, moderate and hard) and duration (shorter, intermediate and longer) by modifying well known paradigms used for clinical assessment of cognitive performance [3739, 41]. The results suggest that fTCD, based on a single presentation of a paradigm, provides questionable results, regarding its feasibility to detect alterations in mild cognitive impairment (MCI) and other conditions. From previous studies in our group, Beishon et al. (2018) demonstrated a significant decrease of CBFV at sustained response during a language task in Alzheimer’s disease but not in MCI when compared with healthy controls [19]. Although the mean of repeated task presentations to each patient might provide more robust estimates of the CBFV changes [83, 84], combining the use of fTCD with other modalities (EEG or near infrared spectroscopy) may enhance discriminatory power in clinical NVC studies [85]. These and other alternatives need to be explored further to provide greater confidence in the use of fTCD in clinical studies of NVC.

There are a number of limitations to our study. First, TCD reflecting CBFV changes in the MCA, can be directly proportional to changes in absolute CBF with the assumption that the diameter of insonated MCA remains constant despite small changes in CO2 and BP [86, 87]. Second, fTCD measurements limited to the MCA might not have the same sensitivity to cognitive stimulation as techniques with greater spatial resolution, such as PET or fMRI [21]. One interesting possibility would be to perform simultaneous measurements in the MCA and PCA as an attempt to improve sensitivity. Third, language task difficulty has been reported to have reduced sensitivity [88, 89], although Payne et al. (2015) showed differences in fTCD responses to the joint fast paced presentation of rhyme and line array judgement tasks [90]. It is possible that the level of difficulty used in our NW paradigms was insufficient to induce differential changes in CBFV responses in the MCA. In addition, the motivation level of participants to engage and perform the tasks to the best of their ability could have affected their CBFV responses, despite our encouragement prior to each test. The role of personal motivation, and how it could be individually assessed, is complex and warrants further investigation [91]. Fourth, the education level could be an important co-variate of cognitive performance. We have not taken this aspect into consideration, given the homogeneity of educational background in our sample, with all participants having had between 16 and 22 years of education. A recent review has highlighted inconsistent findings in previous studies about the effect of ageing on cognitive function [5]. For this reason, further studies are needed to understand the influence of ageing in healthy older adults, and its interaction with the NVC response to tasks with varying complexities and durations. Finally, another limitation was the relatively small sample size (n = 16). However, previous sample size calculations showed that 14 participants could detect a difference of 2% in CBFV at 80% power with p = 0.05, for the case of the P-words task of 60s duration [26]. Although our sample size was approximately the same, it is possible that much larger numbers are needed to detect the subtle differences resulting from NW tasks of greater or lesser complexity in comparison with naming P-words.


We have demonstrated that complexity and duration of NW and SS paradigms provided similar patterns and amplitudes of CBFV responses, without the possibility of making a scalable metric for assessment of NVC in healthy subjects. Nevertheless, some significant changes were seen in comparison with SSREF for the effects of different task durations, indicating a complex interaction between NVC response and duration of stimulation. Further studies, on the sensitivity of the fTCD response to stimulation of varying complexity and duration, is needed by performing similar studies in the influence of ageing, patients with early signs of MCI, Alzheimer’s disease or other types of dementia, exploring alternative cognitive paradigms. Also, a more integrated approach as described previously [22, 24, 92] is needed to take into consideration the potential contribution of other determinants of CBFV, such as BP and EtCO2, to identify parameters of the NVC response that could be scalable, ideally in a linear fashion.


We would like to thank all the participants for their time and effort.


  1. 1. Girouard H, Iadecola C. Neurovascular coupling in the normal brain and in hypertension, stroke, and Alzheimer disease. J Appl Physiol (1985). 2006;100(1):328–35. Epub 2005/12/17. pmid:16357086.
  2. 2. Peterson EC, Wang Z, Britz G. Regulation of cerebral blood flow. Int J Vasc Med. 2011;2011:823525. Epub 2011/08/03. pmid:21808738; PubMed Central PMCID: PMC3144666.
  3. 3. Paulson OB. Blood-brain barrier, brain metabolism and cerebral blood flow. Eur Neuropsychopharmacol. 2002;12(6):495–501. Epub 2002/12/07. pmid:12468012.
  4. 4. Iadecola C. Neurovascular regulation in the normal brain and in Alzheimer’s disease. Nat Rev Neurosci. 2004;5(5):347–60. Epub 2004/04/22. pmid:15100718.
  5. 5. Beishon L, Clough RH, Kadicheeni M, Chithiramohan T, Panerai RB, Haunton VJ, et al. Vascular and haemodynamic issues of brain ageing. Pflugers Arch. 2021;473(5):735–51. Epub 2021/01/13. pmid:33439324.
  6. 6. Logothetis NK, Pauls J, Augath M, Trinath T, Oeltermann A. Neurophysiological investigation of the basis of the fMRI signal. Nature. 2001;412(6843):150–7. Epub 2001/07/13. pmid:11449264.
  7. 7. Hock C, Villringer K, Muller-Spahn F, Wenzel R, Heekeren H, Schuh-Hofer S, et al. Decrease in parietal cerebral hemoglobin oxygenation during performance of a verbal fluency task in patients with Alzheimer’s disease monitored by means of near-infrared spectroscopy (NIRS)—correlation with simultaneous rCBF-PET measurements. Brain Res. 1997;755(2):293–303. Epub 1997/05/02. pmid:9175896.
  8. 8. Goldenberg G, Podreka I, Steiner M, Willmes K, Suess E, Deecke L. Regional cerebral blood flow patterns in visual imagery. Neuropsychologia. 1989;27(5):641–64. Epub 1989/01/01. pmid:2787003.
  9. 9. Jasdzewski G, Strangman G, Wagner J, Kwong KK, Poldrack RA, Boas DA. Differences in the hemodynamic response to event-related motor and visual paradigms as measured by near-infrared spectroscopy. Neuroimage. 2003;20(1):479–88. Epub 2003/10/07. pmid:14527608.
  10. 10. Ances BM, Zarahn E, Greenberg JH, Detre JA. Coupling of neural activation to blood flow in the somatosensory cortex of rats is time-intensity separable, but not linear. J Cereb Blood Flow Metab. 2000;20(6):921–30. Epub 2000/07/14. pmid:10894175.
  11. 11. Aaslid R. Visually evoked dynamic blood flow response of the human cerebral circulation. Stroke. 1987;18(4):771–5. Epub 1987/07/01. pmid:3299883.
  12. 12. Deppe M, Knecht S, Henningsen H, Ringelstein EB. AVERAGE: a Windows program for automated analysis of event related cerebral blood flow. J Neurosci Methods. 1997;75(2):147–54. Epub 1997/08/22. pmid:9288646.
  13. 13. Adams RJ. TCD in sickle cell disease: an important and useful test. Pediatr Radiol. 2005;35(3):229–34. Epub 2005/02/11. pmid:15703904.
  14. 14. Salinet AS, Robinson TG, Panerai RB. Effects of cerebral ischemia on human neurovascular coupling, CO2 reactivity, and dynamic cerebral autoregulation. J Appl Physiol (1985). 2015;118(2):170–7. Epub 2015/01/17. pmid:25593216.
  15. 15. Rigamonti A, Ackery A, Baker AJ. Transcranial Doppler monitoring in subarachnoid hemorrhage: a critical tool in critical care. Can J Anaesth. 2008;55(2):112–23. Epub 2008/02/05. pmid:18245071.
  16. 16. Ducrocq X, Braun M, Debouverie M, Junges C, Hummer M, Vespignani H. Brain death and transcranial Doppler: experience in 130 cases of brain dead patients. J Neurol Sci. 1998;160(1):41–6. Epub 1998/11/06. pmid:9804115.
  17. 17. Pennekamp CW, Moll FL, de Borst GJ. The potential benefits and the role of cerebral monitoring in carotid endarterectomy. Curr Opin Anaesthesiol. 2011;24(6):693–7. Epub 2011/10/06. pmid:21971393.
  18. 18. Thaler D. Patent foramen ovale in older patients with cryptogenic stroke or transient ischaemic attack. Lancet Neurol. 2018;17(7):573–4. Epub 2018/06/12. pmid:29887163.
  19. 19. Beishon LC, Panerai RB, Robinson TG, Subramaniam H, Haunton VJ. The Assessment of Cerebrovascular Response to a Language Task from the Addenbrooke’s Cognitive Examination in Cognitive Impairment: A Feasibility Functional Transcranial Doppler Ultrasonography Study. J Alzheimers Dis Rep. 2018;2(1):153–64. Epub 2018/11/28. pmid:30480258; PubMed Central PMCID: PMC6218154.
  20. 20. Panerai RB. Assessment of cerebral pressure autoregulation in humans—a review of measurement methods. Physiol Meas. 1998;19(3):305–38. Epub 1998/09/15. pmid:9735883.
  21. 21. Stroobant N, Vingerhoets G. Transcranial Doppler ultrasonography monitoring of cerebral hemodynamics during performance of cognitive tasks: a review. Neuropsychol Rev. 2000;10(4):213–31. Epub 2000/12/29. pmid:11132101.
  22. 22. Panerai RB, Moody M, Eames PJ, Potter JF. Cerebral blood flow velocity during mental activation: interpretation with different models of the passive pressure-velocity relationship. J Appl Physiol (1985). 2005;99(6):2352–62. Epub 2005/08/16. pmid:16099892.
  23. 23. Knecht S, Deppe M, Ebner A, Henningsen H, Huber T, Jokeit H, et al. Noninvasive determination of language lateralization by functional transcranial Doppler sonography: a comparison with the Wada test. Stroke. 1998;29(1):82–6. Epub 1998/01/28. pmid:9445333.
  24. 24. Moody M, Panerai RB, Eames PJ, Potter JF. Cerebral and systemic hemodynamic changes during cognitive and motor activation paradigms. Am J Physiol Regul Integr Comp Physiol. 2005;288(6):R1581–8. Epub 2005/01/29. pmid:15677522.
  25. 25. Duschek S, Schandry R. Functional transcranial Doppler sonography as a tool in psychophysiological research. Psychophysiology. 2003;40(3):436–54. Epub 2003/08/30. pmid:12946117.
  26. 26. Beishon LC, Williams CAL, Panerai RB, Robinson TG, Haunton VJ. The assessment of neurovascular coupling with the Addenbrooke’s Cognitive Examination: a functional transcranial Doppler ultrasonographic study. J Neurophysiol. 2018;119(3):1084–94. Epub 2017/12/01. pmid:29187557.
  27. 27. Matteis M, Bivona U, Catani S, Pasqualetti P, Formisano R, Vernieri F, et al. Functional transcranial Doppler assessment of cerebral blood flow velocities changes during attention tasks. Eur J Neurol. 2009;16(1):81–7. Epub 2008/12/18. pmid:19087154.
  28. 28. Montoro CI, Duschek S, Muñoz Ladrón de Guevara C, Fernández-Serrano MJ, Reyes del Paso GA. Aberrant Cerebral Blood Flow Responses During Cognition: Implications for the Understanding of Cognitive Deficits in Fibromyalgia. Neuropsychology. 2015;29(2):173–82. pmid:25151113
  29. 29. Sturzenegger M, Newell DW, Aaslid R. Visually evoked blood flow response assessed by simultaneous two-channel transcranial Doppler using flow velocity averaging. Stroke. 1996;27(12):2256–61. Epub 1996/12/01. pmid:8969790.
  30. 30. Spence EEM, Hodge SVL, Rosentreter R, Lam T, Squair JW, Fisher JP, et al. Visual task complexity and eye movement patterns influence measures of human neurovascular coupling. Physiol Behav. 2021;229:113198. Epub 2020/10/18. pmid:33068563.
  31. 31. Salinet AS, Robinson TG, Panerai RB. Active, passive, and motor imagery paradigms: component analysis to assess neurovascular coupling. J Appl Physiol (1985). 2013;114(10):1406–12. Epub 2013/03/02. pmid:23449939.
  32. 32. Llwyd O, Panerai RB, Robinson TG. Effects of dominant and non-dominant passive arm manoeuvres on the neurovascular coupling response. Eur J Appl Physiol. 2017;117(11):2191–9. Epub 2017/09/07. pmid:28875348.
  33. 33. Burma JS, Wassmuth RM, Kennedy CM, Miutz LN, Newel KT, Carere J, et al. Does task complexity impact the neurovascular coupling response similarly between males and females? Physiol Rep. 2021;9(17):e15020–e. pmid:34514743.
  34. 34. Sorond FA, Kiely DK, Galica A, Moscufo N, Serrador JM, Iloputaife I, et al. Neurovascular coupling is impaired in slow walkers: The MOBILIZE Boston Study. Annals of neurology. 2011;70(2):213–20. pmid:21674588
  35. 35. Csipo T, Lipecz A, Mukli P, Bahadli D, Abdulhussein O, Owens CD, et al. Increased cognitive workload evokes greater neurovascular coupling responses in healthy young adults. PLOS ONE. 2021;16(5):e0250043. pmid:34010279
  36. 36. Sweet LH. N-Back Paradigm. In: Kreutzer JS, DeLuca J, Caplan B, editors. Encyclopedia of Clinical Neuropsychology. New York, NY: Springer New York; 2011. p. 1718–9.
  37. 37. Folstein MF, Folstein SE, McHugh PR. "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. Journal of psychiatric research. 1975;12(3):189–98. Epub 1975/11/01. pmid:1202204.
  38. 38. Mioshi E, Dawson K, Mitchell J, Arnold R, Hodges JR. The Addenbrooke’s Cognitive Examination Revised (ACE-R): a brief cognitive test battery for dementia screening. International Journal of Geriatric Psychiatry. 2006;21(11):1078–85. pmid:16977673
  39. 39. Nasreddine ZS, Phillips NA, Bédirian V, Charbonneau S, Whitehead V, Collin I, et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. Journal of the American Geriatrics Society. 2005;53(4):695–9. Epub 2005/04/09. pmid:15817019.
  40. 40. Beishon L, Williams CAL, Panerai RB, Robinson TG, Haunton VJ. Reproducibility of task activation using the Addenbrooke’s cognitive examination in healthy controls: A functional Transcranial Doppler ultrasonography study. J Neurosci Methods. 2017;291:131–40. Epub 2017/08/23. pmid:28827165.
  41. 41. Williams CAL, Panerai RB, Robinson TG, Haunton VJ. Transcranial Doppler ultrasonography in the assessment of neurovascular coupling responses to cognitive examination in healthy controls: A feasibility study. J Neurosci Methods. 2017;284:57–62. Epub 2017/04/30. pmid:28455103.
  42. 42. Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia. 1971;9(1):97–113. Epub 1971/03/01. pmid:5146491.
  43. 43. Borkowski JG, Benton AL, Spreen O. Word fluency and brain damage. Neuropsychologia. 1967;5(2):135–40.
  44. 44. Kwak SK, Kim JH. Statistical data preparation: management of missing values and outliers. Korean J Anesthesiol. 2017;70(4):407–11. Epub 2017/08/11. pmid:28794835; PubMed Central PMCID: PMC5548942.
  45. 45. Debreczeni R, Amrein I, Kamondi A, Szirmai I. Hypocapnia induced by involuntary hyperventilation during mental arithmetic reduces cerebral blood flow velocity. Tohoku J Exp Med. 2009;217(2):147–54. Epub 2009/02/13. pmid:19212108.
  46. 46. Werner N, Kapan N, Reyes del Paso GA. Patterns of Cerebral Blood Flow and Systemic Hemodynamics During Arithmetic Processing. J Psychophysiol. 2008;22(2):81–90. PubMed PMID: WOS:000267843200003.
  47. 47. Williamson JW. The relevance of central command for the neural cardiovascular control of exercise. Exp Physiol. 2010;95(11):1043–8. Epub 2010/08/11. pmid:20696787; PubMed Central PMCID: PMC3035817.
  48. 48. Boban M, Crnac P, Junakovic A, Malojcic B. Hemodynamic monitoring of middle cerebral arteries during cognitive tasks performance. Psychiatry Clin Neurosci. 2014;68(11):795–803. Epub 2014/04/17. pmid:24735174.
  49. 49. Szirmai I, Amrein I, Palvolgyi L, Debreczeni R, Kamondi A. Correlation between blood flow velocity in the middle cerebral artery and EEG during cognitive effort. Brain Res Cogn Brain Res. 2005;24(1):33–40. Epub 2005/06/01. pmid:15922155.
  50. 50. Sato A, Sato Y, Uchida S. Regulation of regional cerebral blood flow by cholinergic fibers originating in the basal forebrain. Int J Dev Neurosci. 2001;19(3):327–37. Epub 2001/05/05. pmid:11337202.
  51. 51. Feig SL, Guillery RW. Corticothalamic axons contact blood vessels as well as nerve cells in the thalamus. Eur J Neurosci. 2000;12(6):2195–8. Epub 2000/07/11. pmid:10886359.
  52. 52. Fehr T, Code C, Herrmann M. Common brain regions underlying different arithmetic operations as revealed by conjunct fMRI-BOLD activation. Brain Res. 2007;1172:93–102. Epub 2007/09/08. pmid:17822681.
  53. 53. Helton WS, Warm JS, Tripp LD, Matthews G, Parasuraman R, Hancock PA. Cerebral lateralization of vigilance: a function of task difficulty. Neuropsychologia. 2010;48(6):1683–8. Epub 2010/02/23. pmid:20171235.
  54. 54. Fu CH, Morgan K, Suckling J, Williams SC, Andrew C, Vythelingum GN, et al. A functional magnetic resonance imaging study of overt letter verbal fluency using a clustered acquisition sequence: greater anterior cingulate activation with increased task demand. Neuroimage. 2002;17(2):871–9. Epub 2002/10/16. pmid:12377161.
  55. 55. Senhorini MC, Cerqueira CT, Schaufelberger MS, Almeida JC, Amaro E, Sato JR, et al. Brain activity patterns during phonological verbal fluency performance with varying levels of difficulty: a functional magnetic resonance imaging study in Portuguese-speaking healthy individuals. J Clin Exp Neuropsychol. 2011;33(8):864–73. Epub 2011/04/29. pmid:21526446.
  56. 56. Zhang YT, Zhang Q, Zhang J, Li W. Laterality of brain areas associated with arithmetic calculations revealed by functional magnetic resonance imaging. Chin Med J (Engl). 2005;118(8):633–8. Epub 2005/05/19. pmid:15899117.
  57. 57. Kim SJ, Kim IJ, Kim YK, Lee TH, Lee JS, Jun S, et al. Probabilistic anatomic mapping of cerebral blood flow distribution of the middle cerebral artery. J Nucl Med. 2008;49(1):39–43. Epub 2007/12/14. pmid:18077521.
  58. 58. Smirl JD, Wright AD, Bryk K, van Donkelaar P. Where’s Waldo? The utility of a complicated visual search paradigm for transcranial Doppler-based assessments of neurovascular coupling. J Neurosci Methods. 2016;270:92–101. Epub 2016/06/14. pmid:27291357.
  59. 59. Vingerhoets G, Stroobant N. Lateralization of cerebral blood flow velocity changes during cognitive tasks. A simultaneous bilateral transcranial Doppler study. Stroke. 1999;30(10):2152–8. Epub 1999/10/08. pmid:10512921.
  60. 60. Meyer GF, Spray A, Fairlie JE, Uomini NT. Inferring common cognitive mechanisms from brain blood-flow lateralization data: a new methodology for fTCD analysis. Front Psychol. 2014;5:552. Epub 2014/07/02. pmid:24982641; PubMed Central PMCID: PMC4059176.
  61. 61. Rosengarten B, Paulsen S, Molnar S, Kaschel R, Gallhofer B, Kaps M. Activation-flow coupling differentiates between vascular and Alzheimer type of dementia. J Neurol Sci. 2007;257(1–2):149–54. Epub 2007/02/27. pmid:17321550.
  62. 62. Rosengarten B, Osthaus S, Kaps M. Influence of stimulus duration on the neurovascular coupling response. Ultraschall Med. 2004;25(2):116–9. Epub 2004/04/16. pmid:15085452.
  63. 63. Park DC, Reuter-Lorenz P. The adaptive brain: aging and neurocognitive scaffolding. Annu Rev Psychol. 2009;60:173–96. Epub 2008/11/28. pmid:19035823; PubMed Central PMCID: PMC3359129.
  64. 64. Jamadar SD. The CRUNCH model does not account for load-dependent changes in visuospatial working memory in older adults. Neuropsychologia. 2020;142:107446. Epub 2020/04/03. pmid:32234498.
  65. 65. Lisak M, Trkanjec Z, Plavec D, Žigman M, Kusić Z, Kes V, et al. Non-invasive assessment of cerebral blood flow changes during complex activation. Translational Neuroscience. 2012;3(1):28–35.
  66. 66. Deppe M, Knecht S, Papke K, Lohmann H, Fleischer H, Heindel W, et al. Assessment of hemispheric language lateralization: a comparison between fMRI and fTCD. J Cereb Blood Flow Metab. 2000;20(2):263–8. Epub 2000/03/04. pmid:10698062.
  67. 67. Hosford PS, Gourine AV. What is the key mediator of the neurovascular coupling response? Neurosci Biobehav Rev. 2019;96:174–81. Epub 2018/11/28. pmid:30481531; PubMed Central PMCID: PMC6331662.
  68. 68. Duschek S, Hoffmann A, Montoro CI, Bair A, Reyes Del Paso GA, Ettinger U. Cerebral blood flow modulations during antisaccade preparation in chronic hypotension. Psychophysiology. 2019;56(3):e13305. Epub 2018/11/21. pmid:30456801.
  69. 69. Sato K, Moriyama M, Sadamoto T. Influence of central command on cerebral blood flow at the onset of exercise in women. Exp Physiol. 2009;94(11):1139–46. Epub 2009/08/04. pmid:19648481.
  70. 70. Goghari VM, MacDonald AW 3rd. The neural basis of cognitive control: response selection and inhibition. Brain Cogn. 2009;71(2):72–83. Epub 2009/05/12. pmid:19427089; PubMed Central PMCID: PMC2905055.
  71. 71. Schlosser R, Hutchinson M, Joseffer S, Rusinek H, Saarimaki A, Stevenson J, et al. Functional magnetic resonance imaging of human brain activity in a verbal fluency task. J Neurol Neurosurg Psychiatry. 1998;64(4):492–8. Epub 1998/05/12. pmid:9576541; PubMed Central PMCID: PMC2170033.
  72. 72. Vogan VM, Morgan BR, Powell TL, Smith ML, Taylor MJ. The neurodevelopmental differences of increasing verbal working memory demand in children and adults. Dev Cogn Neurosci. 2016;17:19–27. Epub 2015/11/30. pmid:26615571; PubMed Central PMCID: PMC6990091.
  73. 73. Menon V. Developmental cognitive neuroscience of arithmetic: implications for learning and education. ZDM. 2010;42(6):515–25. Epub 2011/10/18. pmid:22003371; PubMed Central PMCID: PMC3193278.
  74. 74. Silvestrini M, Cupini LM, Matteis M, Troisi E, Caltagirone C. Bilateral simultaneous assessment of cerebral flow velocity during mental activity. J Cereb Blood Flow Metab. 1994;14(4):643–8. Epub 1994/07/01. pmid:7912242.
  75. 75. Panerai RB, Salinet AS, Robinson TG. Contribution of arterial blood pressure and PaCO2 to the cerebrovascular responses to motor stimulation. Am J Physiol Heart Circ Physiol. 2012;302(2):H459–66. Epub 2011/11/08. pmid:22058160.
  76. 76. Panczel G, Daffertshofer M, Ries S, Spiegel D, Hennerici M. Age and stimulus dependency of visually evoked cerebral blood flow responses. Stroke. 1999;30(3):619–23. Epub 1999/03/06. pmid:10066861.
  77. 77. Tiecks FP, Haberl RL, Newell DW. Temporal patterns of evoked cerebral blood flow during reading. J Cereb Blood Flow Metab. 1998;18(7):735–41. Epub 1998/07/15. pmid:9663503
  78. 78. Grinband J, Steffener J, Razlighi QR, Stern Y. BOLD neurovascular coupling does not change significantly with normal aging. Hum Brain Mapp. 2017;38(7):3538–51. Epub 2017/04/19. pmid:28419680; PubMed Central PMCID: PMC5882590.
  79. 79. Miller KL, Luh WM, Liu TT, Martinez A, Obata T, Wong EC, et al. Nonlinear temporal dynamics of the cerebral blood flow response. Hum Brain Mapp. 2001;13(1):1–12. Epub 2001/04/03. pmid:11284042; PubMed Central PMCID: PMC6871988.
  80. 80. Carp J, Fitzgerald KD, Taylor SF, Weissman DH. Removing the effect of response time on brain activity reveals developmental differences in conflict processing in the posterior medial prefrontal cortex. Neuroimage. 2012;59(1):853–60. Epub 2011/08/13. pmid:21835249; PubMed Central PMCID: PMC3196273.
  81. 81. Yarkoni T, Barch DM, Gray JR, Conturo TE, Braver TS. BOLD correlates of trial-by-trial reaction time variability in gray and white matter: a multi-study fMRI analysis. PLoS One. 2009;4(1):e4257. Epub 2009/01/24. pmid:19165335; PubMed Central PMCID: PMC2622763.
  82. 82. Settakis G, Lengyel A, Molnar C, Bereczki D, Csiba L, Fulesdi B. Transcranial Doppler study of the cerebral hemodynamic changes during breath-holding and hyperventilation tests. J Neuroimaging. 2002;12(3):252–8. Epub 2002/07/16. pmid:12116744.
  83. 83. Deppe M, Knecht S, Lohmann H, Ringelstein EB. A method for the automated assessment of temporal characteristics of functional hemispheric lateralization by transcranial Doppler sonography. J Neuroimaging. 2004;14(3):226–30. Epub 2004/07/02. pmid:15228762.
  84. 84. Knecht S, Deppe M, Ringelstein EB, Wirtz M, Lohmann H, Drager B, et al. Reproducibility of functional transcranial Doppler sonography in determining hemispheric language lateralization. Stroke. 1998;29(6):1155–9. Epub 1998/06/17. pmid:9626288
  85. 85. Shin J, von Luhmann A, Kim DW, Mehnert J, Hwang HJ, Muller KR. Simultaneous acquisition of EEG and NIRS during cognitive tasks for an open access dataset. Sci Data. 2018;5(1):180003. Epub 2018/02/14. pmid:29437166; PubMed Central PMCID: PMC5810421.
  86. 86. Serrador JM, Picot PA, Rutt BK, Shoemaker JK, Bondar RL. MRI measures of middle cerebral artery diameter in conscious humans during simulated orthostasis. Stroke. 2000;31(7):1672–8. Epub 2000/07/08. pmid:10884472.
  87. 87. Giller CA, Bowman G, Dyer H, Mootz L, Krippner W. Cerebral arterial diameters during changes in blood pressure and carbon dioxide during craniotomy. Neurosurgery. 1993;32(5):737–41; discussion 41–2. Epub 1993/05/01. pmid:8492848.
  88. 88. Badcock NA, Nye A, Bishop DV. Using functional transcranial Doppler ultrasonography to assess language lateralisation: Influence of task and difficulty level. Laterality. 2012;17(6):694–710. Epub 2012/10/27. pmid:23098198; PubMed Central PMCID: PMC3483861.
  89. 89. Rosch RE, Bishop DV, Badcock NA. Lateralised visual attention is unrelated to language lateralisation, and not influenced by task difficulty—a functional transcranial Doppler study. Neuropsychologia. 2012;50(5):810–5. Epub 2012/01/31. pmid:22285903; PubMed Central PMCID: PMC3334833.
  90. 90. Payne H, Gutierrez-Sigut E, Subik J, Woll B, MacSweeney M. Stimulus rate increases lateralisation in linguistic and non-linguistic tasks measured by functional transcranial Doppler sonography. Neuropsychologia. 2015;72:59–69. Epub 2015/04/25. pmid:25908491; PubMed Central PMCID: PMC4922413.
  91. 91. Bendall RCA, Eachus P, Thompson C. A Brief Review of Research Using Near-Infrared Spectroscopy to Measure Activation of the Prefrontal Cortex during Emotional Processing: The Importance of Experimental Design. Frontiers in Human Neuroscience. 2016;10. pmid:27812329
  92. 92. Salinet AS, Robinson TG, Panerai RB. Cerebral blood flow response to neural activation after acute ischemic stroke: a failure of myogenic regulation? J Neurol. 2013;260(10):2588–95. Epub 2013/07/05. pmid:23824356.