P. Mills (UREAP Report)

1

UREAP – Final Report
Investigating age-related cognitive decline with fNIRS and peripheral vascular function analysis
Parker Mills
Primary Supervisor: Dr. Mark Rakobowchuk
Thompson Rivers University
April to September 2024 Term

P. Mills (UREAP Report)

2

1.1 Introduction

Technological and medical advancements as well as an increased standard of living has
resulted in a larger cohort of older adults who are living longer among the population (Cabeza et
al., 2018). Blum et al. (2021), predicts that the population of individuals over the age of 65 is
expected to double over the next two decades. Age is the leading risk factor in the development
of neurodegenerative diseases such as Alzheimer’s and Parkinson’s Disease, and there is strong
evidence that cognitive function declines with age, even in the absence of pathology (Blum et al.,
2021; Cabeza et al., 2018). It is important to understand the reasons and mechanisms behind why
cognitive decline is occurring faster in certain aging individuals whose brains appear healthy
(Cabeza et al., 2018).
Researchers have identified that older adults exhibit reduced lateralization in brain
activity (using one hemisphere or the other) when compared to younger adults, who utilize one
hemisphere while performing the same task (Cabeza 2002). This reduced hemispheric
asymmetry in older adults (HAROLD) is especially apparent in the prefrontal cortex during
working memory tasks. Working memory, which is required for short-term storage and
processing of information for higher level cognition, is limited by both the capacity and ability of
an older individual to activate the required neural regions to accomplish a complex task (Mattay
et al., 2006).
A study by Piefke et al. (2012) showed that when subjected to a visual-spatial working
memory task, older participants had greater prefrontal cortex activation across both hemispheres,
while younger participants primarily activated only the left hemisphere. There are two theories
that offer possible explanations for age-related reduction in hemispheric asymmetry. The first

P. Mills (UREAP Report)

3

theory is the dedifferentiation hypothesis, and the second is the neural compensation theory. The
dedifferentiation theory suggests that brain activity becomes less selective in response to a
stimulus, so brain structures have less processing specificity under specific cognitive loads (Carp
et al., 2010; McDonough et al., 2022; Piefke et al., 2012). Young adults show distinct patterns of
synchronous brain activity within neural networks, whereas these patterns become less
distinguishable in older adults (Grady et al., 2016; McDonough et al., 2022). Conversely, neuralcompensation refers to the increase in brain activity to enhance cognitive performance through
either upregulation, selection, or reorganization (Bunzeck et al., 2023; McDonough et al., 2022).
The compensation-related utilization of neural circuits hypothesis (CRUNCH), the primary
model used to test compensation, describes a quadratic function in regard to brain activation
under increasing cognitive loads. When the task is easy, there is limited brain activation in
younger and older adults, but as the task becomes more difficult, brain activity increases until it
reaches a threshold where the task is too difficult and brain activity decreases (Lorenz and
Cappell 2008). The difference between younger and older adults is that compensatory brain
activity in older adults is expected to be at its maximum during lower task demands (Lorenz and
Park 2014).
Carp et al. (2010) tested CRUNCH by having younger (18-25yrs) and older (61-82 yrs)
adults complete working memory tasks while measuring brain activity patterns using functional
magnetic resonance imaging (fMRI). First, they found that during memory encoding and
retrieval, older adults had reduced distinctiveness regardless of memory load. Second, during
memory maintenance at lower task loads, older adults had higher activation of PFC regions than
young adults, but as the memory load increased, the older adults had reduced distinctiveness and
increased recruitment of neural resources until they eventually hit a limit and could no longer

P. Mills (UREAP Report)

4

perform the task (Carp et al., 2010). Although these results support CRUNCH and thus,
compensation, they also provided evidence for dedifferentiation. Additionally, several
researchers have provided evidence against CRUNCH, with older adults being able to show
compensation even in high cognitive load tasks (Jamadar 2020; Ranchod et al., 2023). During a
picture memory encoding task, Duzel et al. (2011), found a subgroup of older adults who,
without compensatory activity, had recollection memory performance that was indistinguishable
from the young adult cohort. However, there were still older adults in the study who exhibited
compensatory activity in the prefrontal and parietal regions to maintain memory performance
(Duzel et al., 2011).

1.2 Using fNIRS to Test CRUNCH and Measure Systemic Vascular Function

Functional near-infrared spectroscopy (fNIRS) is a portable and versatile neuroimaging
tool that can be used to measure brain activity by observing the hemodynamic response in the
cortical tissue of the brain. Neural activity is associated with an increase in regional arteriolar
vasodilation caused by a larger demand in nutrients, such as glucose and oxygen, from local
neurons which causes an influx of oxygenated haemoglobin (Pinti et al., 2018). The changes in
concentration of oxygenated and deoxygenated haemoglobin can then be detected when looking
at the differences in absorption of near-infrared light in the cortical layer of the brain (Pinti et al.,
2018). It is imperative that the CRUNCH hypothesis be more robustly tested using our fNIRS
equipment, so that we can examine the possibility that the compensation effect is related to the
working memory task itself, and not just a general mechanism to increase neurocognitive
functionality (Jamadar 2020). Therefore, it is possible to incorporate a Corsi-block tapping task

P. Mills (UREAP Report)

5

(rather than a N-back task) that involves a 9 square array where the length of the sequence of the
visual stimulus can be modified to vary task difficulty (Brunetti et al., 2014). The hemodynamic
observations made in the prefrontal cortex during a Corsi-block tapping task of varying
complexity will help give us a better understanding of the brain activity pattern differences
between older and younger adults and how neural compensation is involved in cognitive function
as we age.
A review by Toth et al. (2017) sates having adequate nutrient and oxygen supply through
the cerebromicrovascular network is critical in the maintenance of normal brain function. Agerelated dysregulation of the cerebral blood flow (CBF) network is implicated in the advancement
of cognitive decline, namely in the form of dementia. Autoregulation of CBF from both larger
proximal arteries and the microvasculature of the brain is imperative to prevent both
hyperperfusion and hypoperfusion for optimal oxygen and nutrient transport to neurons. As
people age, their arteries become less elastic, causing high hemodynamic pulsatility. This
hypertensive action may increase blood pressure in the cerebral microcirculation causing
cerebral brain capillary damage and, consequently, cognitive impairment (Toth et al., 2017).
Toth et al. (2017) also explain studies on aging concluded that endothelial dysfunction
increases because of the accumulation of reactive oxygen species that decrease the
bioavailability of nitric oxide (NO): a key local mediator in vasodilation. This general
mechanism of endothelial dysfunction is likely to contribute to cerebral hypoperfusion,
exacerbating cognitive decline (Toth et al. 2017).
Moreover, there is little research measuring the function of blood vessels in the periphery
of the cardiovascular system as a potential explanation for compensatory brain mechanisms
(Ransom et al., 2024). Owens et al. (2022) investigated the correlation between microvascular

P. Mills (UREAP Report)

6

endothelial dysfunction in the periphery and its effect on neurovascular coupling and thus
cognitive performance during a N-back working memory task. Normal brain function is
dependent on appropriate adjustments to cerebral oxygenated blood supply, which is in turn
dependent on effective microvascular endothelial function.
Owens et. al (2022) enlisted participants with peripheral artery disease as well as healthy
participants to use as a control group for the measurements. Participants’ brain activity was
observed in the prefrontal and motor cortex regions using an fNIRS cap and optode array. Their
endothelial function was measured using the flow-mediated dilation (FMD) technique. The study
found that individuals with peripheral artery disease had decreased neurovascular coupling and
general endothelial microvascular dysfunction, which could be correlated with a decrease in
cognitive task performance (Owens et. al 2022). It is therefore important to assess peripheral
vascular function in tandem with compensatory hemodynamic mechanisms observed in the
brains of aging individuals to better understand the changes in neurocognitive patterns. For the
purpose of this report I will be focusing on the methods and preliminary results for
measurements pertaining to microvascular function and arterial function.

2. Methods: Measuring Endothelial Function Using NIRS and Its Relationship to FlowMediated Dilation

2.1 Participants

This completed study will involve 40 total participants consisting of 20 young
participants (18-25 years old), and 20 older participants (>65 years old). The younger

P. Mills (UREAP Report)

7

participants have been recruited by word of mouth referrals and will be continually recruited
through posters placed on bulletin boards around TRU, the TCC, and local restaurants and cafes.
The older adults will be recruited by contacting local retirement homes, community centers, and
through social media such as Facebook. Participants were given a consent form to read through
and sign explaining the experiment, the goals of the experiment, how their information will be
used, and what is expected from them. The consent form also contains exclusion criteria to
determine participant eligibility, this includes: the aforementioned age restrictions, requiring
good vision, English language proficiency, having no known neurological complications
(traumatic brain injury, stroke, mild cognitive impairment) or psychological disorders
(depression, bipolar disorder), right-handedness, six years of formal education, and no attention
enhancing medications or psychoactive drugs. Along with the consent form, Older participants
will also be required to complete a Montreal Cognitive Assessment (MoCA) to determine overall
cognitive function.

2.2 Procedure and Analysis

Initial phases prior to testing first involved assessing and validating techniques to assess
vascular function using muscle NIRS not previously available at TRU. As such, the initial 6
weeks of the UREAP was dedicated to obtaining pilot data and examining the output and optimal
procedures using the MOXY muscle oxygenation monitor (Fortiori Design LLC, Minnesota).
This device was used to obtain oxygen saturation measurements in the microvasculature of the
forearm. The monitor has one emitter that propagates near-infrared light within the range of 630
to 850 nm into the tissue which is then absorbed at different wavelengths by oxygenated and

P. Mills (UREAP Report)

8

deoxygenated hemoglobin. Two detectors, one spaced at 12.5 and the other at 25 mm from the
emitter, then simultaneously sample the light intensity to calculate oxygen saturation of the
underlying microvasculature. Based upon previous studies by McLay et al. (2016) data collection
required the device to be set at a sampling rate of 1Hz to enable adequate temporal resolution to
determine the 10 second reperfusion slope as recommended by McLay et al. (2016), which will
be discussed later. The Moxy monitor unit was connected via Bluetooth to the VO2 Master
mobile app (version 0.41.1), so that the data from the unit could be downloaded and exported
into excel as a .csv file.
For this pilot phase, participants were asked to refrain from exercise, caffeine, and
alcohol for a period of 24 hours prior to testing and to be in a fasted state. The participants were
required to rest for 20 minutes in a supine position before three measurements of blood pressure
were taken by an automated cuff. Participant height, weight and sex was also accounted for in
the MOXY monitor settings. A blood pressure cuff was placed on the upper arm above the elbow.
The muscle oxygenation monitor was then placed at a distance one-third of the way down the
inner forearm from the elbow to the wrist. The device was also wrapped with a tensor bandage to
block any external light from being detected (Rogers et al., 2022; Soares et al. 2020). To test the
reproducibility of the NIRS muscle oxygenation experimental design, two vascular occlusion
tests were administered on four individuals on two separate days. The coefficient of variation
was 25.94% ± 16.02% (n=4). High variation could be partially because after 5 minutes of
occlusion, the sensor would stop collecting data if the oxygen saturation percentage dropped to
0%. This was followed by a horizontal lag-phase in the reperfusion slope analysis. During these
pilot tests the device was only sampling every 2 seconds, so this is why we increased the
sampling rate to 1Hz (McLay et al., 2016).

P. Mills (UREAP Report)

9

Since future tests in the young and older adult populations were to be performed nonsimultaneously, the vascular occlusion test (VOT) using NIRS and FMD described by Soares et
al. (2020) was used. However, this was modified so that the NIRS-VOT was done on the right
arm and the FMD-VOT was done on the left arm. The MOXY monitor recorded a baseline
hemodynamic response at rest for at least 10 minutes. The blood pressure cuff was inflated past
200 mmHg, which is beyond systolic blood pressure to occlude the brachial artery and induce
ischemia. Occlusion lasted for 3 minutes, during which the oxygen saturation in the muscle
tissue dropped. A 3-minute occlusion period was used instead of a 5-minute occlusion period
because the sensor would stop recording data if oxygen saturation dropped to 0.0% (Bezemer et
al., 2009). After 3-minutes the pressure cuff was released and recording continued until the
percent oxygen saturation values returned to baseline. The data was then used to calculate the
10s reperfusion slope, to evaluate microvascular endothelial function (Soares et al., 2020).
The FMD-VOT was also assessed for reproducibility since operator error can add to
variability in the measure. This technique involves imaging vascular ultrasound that requires
substantial practice to gain expertise. The UREAP period was also used to acquire this training.
An EPIQ 5 ultrasound machine (Philips, Professional Healthcare, The Netherlands) with an L123 probe was used to measure the left brachial artery at rest in participants on two occasions
separated by at least several days. The participants were connected to an electrocardiograph
(ECG), so that heart rate could be used to average the arterial diameter data over three heart
cycles and smooth the data. A 30s video clip was acquired for a baseline measurement of the
artery. A blood pressure cuff was positioned around the forearm and inflated to a pressure of
200mmHg (far above systolic blood pressure). After an ischemia period of 5 minutes, the probe
was placed on the arm in the same position that the baseline brachial artery image was acquired,

P. Mills (UREAP Report)

10

and a new video clip was acquired while the cuff was rapidly deflated. Thirty second clips were
then acquired over the next 2.5 minutes to enable the peak diameter of the artery to be obtained.
The percent difference between the baseline and peak artery diameter is typically used as an
indicator of peripheral macrovascular endothelial function. An ultrasound image analysis
software (CAROLAB, IEEE Ultrasonic Symposium, Glasgow, Scotland, version 5.0) was used
to measure the artery diameter for each frame (23fps) of the image. The data was exported to
excel along with the pulse, blood velocity, and ultrasound audio signals for further analysis.

3. Results

Initially, ultrasound imaging accuracy was evaluated using participant FMD samples and
calculating the percent difference between day one and day two measurements. Refer to Table 1
to see coefficient of variation measurements.
Table 1. Comparing percent difference between day one and day two trials to verify the precision
of FMD measurements over two days (n=4).
Participant
Trial 1 % change
Trial 2 % change
Coefficient of
Variation (%)
1
4.73
6.07
17.51
2
9.71
5.22
42.44
3
10.72
10.59
0.87
4
7.99
6.51
14.37
Mean 18.80
Standard Deviation ±17.34
The day to day variation in FMD measurements had a mean percent coefficient of 18.80%
(±17.34%) which is marginally high. The day-to-day variation indicates poor precision.
However, the day-to-day measurements for participant 3 were very precise indicating good
precision for that sample. Using another analysis technique, we could calculate the sum of the

P. Mills (UREAP Report)

11

absolute minimum arterial diameter and the absolute maximum arterial diameter to present a
complete vascular response. Table 2 shows the coefficient of variation results using this different
analysis technique.
Table 2. Comparing percent difference between day one and day two trials to verify the precision
of measurements using the total vascular response (n=4).
Participant
Trial 1 % change
Trial 2 % change
Coefficient of
Variation (%)
1
6.68
6.44
2.62
2
8.45
16.50
45.67
3
10.72
11.66
5.95
4
14.47
11.85
14.07
Mean 17.08
Standard Deviation ±19.66
The mean total vascular response coefficient of variation is 17.08% (±19.66%) which is more
precise than the mean FMD variation values. The improved precision can indicate that the period
of rest prior to collecting a baseline clip was insufficient, so moving forward we increased the
rest period to at least 30 minutes. After testing the precision of our testing method, we started to
collect FMD data in the brachial artery to measure the peripheral vascular function of our
participants. Table 3 features the percent change calculated for each participant’s baseline and
peak brachial artery diameters before and after the 5 minute occlusion.
Table 3. Percent change in each participant’s brachial artery diameter (mm) pre and post cuff
occlusion (n=5).
Diameter (mm)
Participant
Baseline
Peak
% change
1
4595
4796
4.37
2
3796
4053
6.77
3
4031
4297
6.60
4
3584
3781
5.50
5
3323
3452
3.88
Mean
5.42
Standard Deviation
±1.29

P. Mills (UREAP Report)

12

The mean percent change in brachial arterial diameter was 5.42% with a standard deviation of
±1.29%. These FMD percent change values will be compared to literature FMD values for
young, healthy adults in the discussion. Figure 1 shows a boxplot comparison of the baseline
values for brachial artery diameter (mm) to the peak values for brachial artery diameter in
response to cuff release after a 5 minute period of ischemia.

Figure 1. Comparing the mean baseline brachial artery diameter (mm) before cuff release to the
mean peak brachial artery diameter (mm) after a 5 minute occlusion (n=5). The diagonal lines
connect individual participant data.
We can then needed to calculate the 10 second reperfusion slope, in percent oxygen saturation
per second (%O2/sec) from the NIRS-VOT, so that we could compare these values to

P. Mills (UREAP Report)
participants’ FMD responses. Table 4 reports the 10 second reperfusion slopes for our
participants combined with the previously shown FMD responses (% change).
Table 4. The 10 second reperfusion slope (%O2/sec) recorded from the forearm of each
participant after a 3 minute ischemia compared to their FMD response (n=5).
Participant
Slope (%O2/sec)
FMD Response (% change)
1
3.0
4.37
2
1.7
6.77
3
1.5
6.60
4
3.0
5.50
5
5.6
3.88
Mean
2.96
The mean 10 second reperfusion slope was 2.96 %O2/sec. Each participants’ slope value was
then graphed in relationship with their FMD response value to observe the correlative
relationship. Figure 2 examines how the 10 second reperfusion slope and FMD response are
correlated.

13

P. Mills (UREAP Report)

14

FMD Response (% change from baseline to peak)

8
y = -0.7094x + 7.5124
R² = 0.801

7
6
5
4
3
2
1
0
0

1

2
3
4
10 Second Reperfusion Slope (%O2/sec)

5

6

Figure 2. The relationship between the FMD response (% change of arterial diameter from
baseline to peak) and the 10 second reperfusion slope (%O2/sec) after vascular occlusion
(n=5). The slope is negative and the correlation is 0.8.
The participants’ FMD responses and the 10 second reperfusion slopes are highly correlated (R2
of 0.8). The slope is negative indicating that as the reperfusion slope gets steeper the FMD
response should be weaker.

4. Discussion

The preliminary FMD results are promising. They reveal that arterial dilation is occurring
and being measured during the vascular occlusion test. As expected participant’s brachial arteries
are responding to the 5 minute occlusion period and subsequent release by increasing in diameter

P. Mills (UREAP Report)

15

to accommodate the increased pressure experienced along the endothelial lining. A meta-analysis
incorporating 88 clinical trials by Heiss et al. (2022) found that the FMD values in healthy
individuals was 6.4% (95% CI 6.2-6.7%). Our mean FMD value was 5.4% which seems
reasonable given that it is within the range of Heiss et al.’s 6.4% mean from a revision of clinical
trials, and we can’t assume all of our participants have equivalent arterial health. Heiss et al. also
suggest that baseline brachial artery diameter does have a significant impact (-0.44%/mm) in
FMD values. Furthermore, the FMD values for participants 2 and 3 do align with what is
expected for healthy individuals, and participant 4 is approaching the expected range for healthy
individuals. Participant 5’s FMD value can potentially be explained by the effect of having a
smaller brachial artery diameter at rest compared to other participants. Participant 1’s FMD value
is surprising, however statistical outliers are expected to appear within the data.
Our precision measurements obtained before continuing with the experiment were not
very consistent or precise, but this is not critical because in our study we are only taking one
FMD measurement for each participant on the same day. The important part is that the post-cuff
release ultrasound image is taken in the same area as the baseline ultrasound image so that the
analysis is accurate. It is also crucial to collect clear ultrasound images with defined boarders, so
the CAROLAB ultrasound image software can accurately trace the borders of the artery.
The strong correlational relationship between the FMD response and the 10 second
reperfusion slope analysis does align with Soares et al. (2019), however the direction of the
relationship in our study is negative. It is expected that as one variable increases as should the
other in tandem. It is not possible to claim whether these preliminary results are significant
because of the small sample size. Therefore increasing the sample size will help us further
examine this relationship’s magnitude and direction.

P. Mills (UREAP Report)

16

5. Future Directions

This project will be continued as an honours project, so more data will be collected in the
Fall 2024 and Winter 2024 semester. It is imperative to collect enough quality data from young,
healthy participants to be able to observe the correlation between healthy endothelial function
and good cognitive function. It is expected that older adults will have lower FMD values because
endothelial function declines with age. The same observations between young and old adults is
expected to be reflected in NIRS-VOT 10 second reperfusion slope analyses, where older adults
are predicted to have smaller slopes as a result of age-related decline in endothelial function.
The fNIRS testing will examine brain activity in both young and old adults. We
hypothesize that the older adults will generally show increased bilateral brain activity during the
working memory task when compared to young adults. Older adults with lower cognitive
function (identified by low MoCA scores) are predicted to increase bilateral compensation during
lower complexity tasks. Altering task complexity will also allow us to investigate CRUNCH,
which predicts that older adults will exhibit compensatory brain activity up until the task
becomes too difficult at which point brain activity is expected to decrease. Lastly, the Corsiblock tapping task enables us to verify that the neural compensation effect is not simply task
specific but a general method for older individuals to increase cognitive function. Lastly, the
fNIRS data can be analyzed with the data from the two vascular occlusion tests to see if there are
correlations between poor systemic endothelial function and decreased brain efficiency and
lateralization.

P. Mills (UREAP Report)

17

6. References

Bezemer, R., Lima, A., Myers, D., Klijn, E., Heger, M., Goedhart, P. T., Bakker, J., & Ince, C.
(2009). Assessment of tissue oxygen saturation during a vascular occlusion test using nearinfrared spectroscopy: The role of probe spacing and measurement site studied in healthy
volunteers. Critical Care, 13(S5), S4. https://doi.org/10.1186/cc8002

Blum, L., Rosenbaum, D., Röben, B., Dehnen, K., Maetzler, W., Suenkel, U., Fallgatter, A. J.,
Ehlis, A.-C., & Metzger, F. G. (2021). Age-related deterioration of performance and increase of
cortex activity comparing time- versus item-controlled fNIRS measurement. Scientific Reports,
11(1), 6766. https://doi.org/10.1038/s41598-021-85762-w

Brunetti, R., Del Gatto, C., & Delogu, F. (2014). Ecorsi: Implementation and testing of the corsi
block-tapping task for digital tablets. Frontiers in Psychology, 5.
https://doi.org/10.3389/fpsyg.2014.00939

Bunzeck, N., Steiger, T. K., Krämer, U. M., Luedtke, K., Marshall, L., Obleser, J., & Tune, S.
(2024). Trajectories and contributing factors of neural compensation in healthy and pathological
aging. Neuroscience & Biobehavioral Reviews, 156, 105489.
https://doi.org/10.1016/j.neubiorev.2023.105489

Cabeza, R. (2002). Hemispheric asymmetry reduction in older adults: The HAROLD model.
Psychology and Aging, 17(1), 85–100. doi: 10.1037/0882-7974.17.1.85 4

P. Mills (UREAP Report)

18

Cabeza, R., Albert, M., Belleville, S., Craik, F. I. M., Duarte, A., Grady, C. L., Lindenberger, U.,
Nyberg, L., Park, D. C., Reuter-Lorenz, P. A., Rugg, M. D., Steffener, J., & Rajah, M. N. (2018).
Maintenance, reserve and compensation: The cognitive neuroscience of healthy ageing. Nature
Reviews Neuroscience, 19(11), 701–710. https://doi.org/10.1038/s41583-018-0068-2

Carp, J., Gmeindl, L., & Reuter-Lorenz, P. A. (2010). Age differences in the neural
representation of working memory revealed by Multi-Voxel Pattern Analysis. Frontiers in
Human Neuroscience, 4. Available from https://doi.org/10.3389/fnhum.2010.00217

Düzel, E., Schütze, H., Yonelinas, A. P., & Heinze, H. (2011). Functional phenotyping of
successful aging in long‐term memory: Preserved performance in the absence of neural
compensation. Hippocampus, 21(8), 803–814. https://doi.org/10.1002/hipo.20834

Grady, C., Sarraf, S., Saverino, C., & Campbell, K. (2016). Age differences in the functional
interactions among the default, frontoparietal control, and dorsal attention networks.
Neurobiology of Aging, 41, 159–172. https://doi.org/10.1016/j.neurobiolaging.2016.02.020

Heiss, C., Rodriguez-Mateos, A., Bapir, M., Skene, S. S., Sies, H., & Kelm, M. (2023). Flowmediated dilation reference values for evaluation of endothelial function and cardiovascular
health. Cardiovascular Research, 119(1), 283–293. https://doi.org/10.1093/cvr/cvac095

P. Mills (UREAP Report)

19

Jamadar, S. D. (2020). The CRUNCH model does not account for load -dependent changes in
visuospatial working memory in older adults. Neuropsychologia, 142. doi:
10.1016/j.neuropsychologia.2020.107446

Mattay, V. S., Fera, F., Tessitore, A., Hariri, A. R., Berman, K. F., Das, S., Meyer-Lindberg, A.,
Goldberg, T.E., Callicott, J.H., Weinberger, D. R. (2006). Neurophysiological correlates of agerelated changes in working memory capacity. Neuroscience Letter, 392, 32-37. doi:
10.1016/j.neulet.2005.09.025

McDonough, I. M., Nolin, S. A., Visscher, K. M. (2022). 25 years of neurocognitive aging
theories: What have we learned?. Frontiers in Aging Neuroscience, 14. doi:
10.3389/fnagi.2022.1002096

McLay, K. M., Fontana, F. Y., Nederveen, J. P., Guida, F. F., Paterson, D. H., Pogliaghi, S., &
Murias, J. M. (2016). Vascular responsiveness determined by near‐infrared spectroscopy
measures of oxygen saturation. Experimental Physiology, 101(1), 34–40.
https://doi.org/10.1113/EP085406

Owens, C. D., Mukli, P., Csipo, T., Lipecz, A., Silva-Palacios, F., Dasari, T. W., Tarantini, S.,
Gardner, A. W., Montgomery, P. S., Waldstein, S. R., Kellawan, J. M., Nyul-Toth, A.,
Balasubramanian, P., Sotonyi, P., Csiszar, A., Ungvari, Z., & Yabluchanskiy, A. (2022).
Microvascular dysfunction and neurovascular uncoupling are exacerbated in peripheral artery
disease, increasing the risk of cognitive decline in older adults. American Journal of Physiology-

P. Mills (UREAP Report)

20

Heart and Circulatory Physiology, 322(6), H924–H935.
https://doi.org/10.1152/ajpheart.00616.2021

Pinti, P., Tachtsidis, I., Hamilton, A., Hirsch, J., Aichelburg, C., Gilbert, S., & Burgess, P. W.
(2020). The present and future use of functional near‐infrared spectroscopy (fNIRS) for cognitive
neuroscience. Annals of the New York Academy of Sciences, 1464(1), 5–29.
https://doi.org/10.1111/nyas.13948

Piefke, M., Onur, Ö. A., & Fink, G. R. (2012). Aging-related changes of neural mechanisms
underlying visual-spatial working memory. Neurobiology of Aging, 33(7), 1284–1297.
https://doi.org/10.1016/j.neurobiolaging.2010.10.014

Ranchod, S., Rakobowchuk, M., Gonzalez, C. (2023). Distinct age-related brain activity patterns
in the prefrontal cortex when increasing cognitive load: A functional near-infrared spectroscopy
study. PLoS ONE, 18(12). doi: 10.1371/journal.pone.0293394

Ransom, K. V., Traylor, M. K., Batman, G. B., Mulekar, M. S., Hill, B. D., Nelson, A. R., Keller,
J. L. (2024). Arterial stiffness mediates the association between age and processing speed at low
levels of microvascular function in humans across the adult lifespan. American Journal of
Physiology Heart and Circulatory Physiology, 326, 346- 356. doi: 10.1152/ajpheart.00662.2023

P. Mills (UREAP Report)

21

Reuter-Lorenz, P. A., & Cappell, K. A. (2008). Neurocognitive aging and the compensation
hypothesis. Current Directions in Psychological Science, 17(3), 177–182.
https://doi.org/10.1111/j.1467-8721.2008.00570.x

Reuter-Lorenz, P. A., Park, D. C. (2014). How Does it STAC Up? Revisiting The Scaffolding
Theory of Aging and Cognition. Neuropsychology Review, 24(3), 355-370. doi: 10.1007/s11065014-9270-9

Rogers, E. M., Banks, N. F., & Jenkins, N. D. M. (2023). Metabolic and microvascular function
assessed using near‐infrared spectroscopy with vascular occlusion in women: Age differences
and reliability. Experimental Physiology, 108(1), 123–134. https://doi.org/10.1113/EP090540

Soares, R. N., De Oliveira, G. V., Alvares, T. S., & Murias, J. M. (2020). The effects of the
analysis strategy on the correlation between the NIRS reperfusion measures and the FMD
response. Microvascular Research, 127, 103922. https://doi.org/10.1016/j.mvr.2019.103922

Tan, Q., Wang, Y., Li, Z., Wang, D., Lam, W.-K., Wong, D. W.-C., Peng, Y., Zhang, G., & Zhang,
M. (2021). Spectral analysis of muscle hemodynamic responses in post-exercise recovery based
on near-infrared spectroscopy. Sensors, 21(9), 3072. https://doi.org/10.3390/s21093072

Toth, P., Tarantini, S., Csiszar, A., & Ungvari, Z. (2017). Functional vascular contributions to
cognitive impairment and dementia: Mechanisms and consequences of cerebral autoregulatory
dysfunction, endothelial impairment, and neurovascular uncoupling in aging. American Journal

P. Mills (UREAP Report)
of Physiology-Heart and Circulatory Physiology, 312(1), H1–H20.
https://doi.org/10.1152/ajpheart.00581.2016

22