Draft · Draft The effects of aerobic exercise intensity on memory in older adults Journal: Applied...

Draft

The effects of aerobic exercise intensity on memory in older adults

Journal: Applied Physiology, Nutrition, and Metabolism

Manuscript ID apnm-2019-0495.R1

Manuscript Type: Article

Date Submitted by the Author: 23-Sep-2019

Complete List of Authors: Kovacevic, Ana; McMaster University Faculty of ScienceFenesi, Barbara; McMaster University Faculty of SciencePaolucci, Emily; McMaster University Faculty of ScienceHeisz, Jennifer; McMaster University Faculty of Science

Novelty bullets: points that summarize the key findings in

the work:

• High-intensity interval training results in the greatest memory performance in inactive older adults, • Improvement in fitness correlates with improvement in memory performance

Keyword: Memory, Executive functions, Cognition, exercise < exercise, physical activity < exercise, High-intensity interval training, aging < aging, BDNF

Is the invited manuscript for consideration in a Special

Issue? :Not applicable (regular submission)

https://mc06.manuscriptcentral.com/apnm-pubs

Applied Physiology, Nutrition, and Metabolism

Draft

1

Running head: EXERICSE INTENSITY AND COGNITION IN AGING

The effects of aerobic exercise intensity on memory in older adults

Ana Kovacevica, Barbara Fenesib, Emily Paoluccia, Jennifer J. Heisz

McMaster University

Author affiliations

All authors were, at the time of the study, affiliated with Department of Kinesiology,

McMaster University, 1280 Main Street West, Hamilton, Ontario, Canada, L8S 4K1.

Correspondence concerning this article should be directed to Jennifer J. Heisz, Ph.D.,

Associate Professor, Department of Kinesiology, McMaster University, 1280 Main Street West

Hamilton, Ontario, Canada, L8S 4K1, www.neurofitlab.com, [email protected].

Ana Kovacevic, [email protected]; Barbara Fenesi, [email protected]; Emily

Paolucci, [email protected]; Jennifer J. Heisz, [email protected].

a Ana Kovacevic and Emily Paolucci are now at the Faculty of Medicine, University of Toronto, Toronto, Canada. b Barbara Fenesi is now at the Faculty of Education, Western University, London, Canada.

Page 1 of 38



Draft

EXERCISE INTENSITY AND COGNITION IN AGING 2

Abstract

Aerobic exercise may enhance memory in older adults. However, the optimal intensity

and underlying mechanism are unclear. This community-based study examined the effect of

aerobic exercise intensity on memory and general cognitive abilities. Brain-derived neurotrophic

factor (BDNF) was examined as a potential mechanism. Sixty-four sedentary older adults

participated in one of three groups: 1) high-intensity interval training (HIIT); 2) moderate

continuous training (MCT); or 3) stretching control (CON). Prior to and following the

intervention, high-interference memory was assessed using a Mnemonic Similarity task and

executive functions were assessed using Go Nogo and Flanker tasks. HIIT led to the greatest

memory performance compared to MCT and CON [F(2, 55) = 6.04, p = .004] and greater

improvements in memory correlated with greater increases in fitness [rs(46) = .27, p = .03].

Exercise intensity seemed to matter less for executive functioning, as positive trends were

observed for both HIIT and MCT. No significant differences in BDNF were found between

groups. Overall, these results suggest that aerobic exercise may enhance memory in older adults,

with the potential for higher intensity exercise to yield the greatest benefit. While our findings

suggest that BDNF does not regulate these adaptations, the mechanisms remain to be

determined.

Novelty bullets:

High-intensity interval training results in the greatest memory performance in inactive

older adults compared to moderate continuous training or stretching

Improvement in fitness correlates with improvement in memory performance

Keywords: Memory; Executive functions; Cognition, Exercise, Physical activity; High-

intensity interval training; Aging; BDNF

Page 2 of 38



Draft


Introduction

Aging is associated with a progressive decline in memory (Fjell et al. 2014; Scullin and

Bliwise 2015) and physical exercise has been identified as a potentially therapeutic treatment to

mitigate this decline (Bherer et al. 2013; Zheng et al. 2016). Higher intensity exercise improves

memory for younger adults (Bherer et al. 2013; Déry et al. 2013; Heisz et al. 2017; Heisz et al.

2015); however, the optimal exercise intensity to improve memory in older adults is unknown.

The present study examined the impact of different aerobic exercise intensities on memory and

other cognitive abilities in older adults.

Advancing age is associated with impaired cognitive function due to the atrophy of key

brain structures including the hippocampus, prefrontal cortex (Erickson et al. 2011; Fjell et al.

2014) and white matter tracts (Daselaar et al. 2013). High-interference memory is a subtype of

memory function that is particularly vulnerable to age-related changes (Stark et al. 2015). This

type of memory requires the creation of distinct, non-overlapping representations to keep similar

items separate and minimize interference (Yassa and Stark 2011). Prior work suggests that older

adults require a greater degree of dissimilarity between inputs to encode them as distinct stimuli,

a phenomenon termed “representational rigidity” (Bullock et al. 2018; Stark et al. 2013). Thus,

older adults tend to recruit pattern completion processes, such that they have a propensity to

reactivate previously encoded representations when presented with highly similar inputs (Yassa

and Stark 2011). This age-related decline in high-interference memory makes it more difficult

for older adults to retrieve relevant information in similar situations; critically, this can

compromise decision-making (e.g., recalling whether the last dose of medication was taken

today or yesterday) and social interactions (e.g., recognizing what information has already been

conveyed to a person).

Page 3 of 38



Draft


Although memory naturally declines with age, exercise may have the potential to slow

the rate of decline (Bherer et al. 2013). Specifically, exercise mitigates age-related volume loss

in brain regions implicated in memory function (Erickson et al. 2011) and enhances the

functional connectivity of the white matter tracts to improve signal transduction (Fjell et al.

2014). These structural fortifications from exercise translate into functional improvements; for

example, greater hippocampal cerebral blood volume is associated with better spatial memory

(Erickson et al. 2011; Hillman et al. 2008; Honea et al. 2009), greater white matter integrity is

associated with faster speed of information processing (Papp et al. 2014; Voss et al. 2010), and

greater prefrontal cortical volume is associated with better general cognitive abilities such as

inhibitory control, processing speed, and visuospatial processing (Colcombe et al. 2006).

Exercise also increases the production of new neurons in the hippocampus in animal

models to bolster cognitive functions governed by the hippocampus, including learning and

memory (Curlik and Shors 2013). One peripheral indicator of neurogenesis is brain derived

neurotrophic factor (BDNF), which helps to support the growth and survival of neurons (Cotman

and Berchtold 2002). Evidence from animal models suggests that BDNF is particularly

important for high-interference memory, as it facilitates hippocampal neurogenesis (Bekinschtein

et al. 2011). Aerobic exercise increases both BDNF protein and mRNA in the hippocampus of

aged rats, and these increases are correlated with improved memory function (Fu et al. 2017).

Similar findings are observed in younger human participants, with exercise increasing peripheral

BDNF and high-interference memory (Déry et al. 2013). Likewise, in older humans participants,

an acute bout of exercise can increase peripheral BDNF (Coelho et al. 2014); however, the

impact of exercise and BDNF on high-interference memory for older adults has yet to be

established.

Page 4 of 38



Draft


The optimal intensity of aerobic exercise needed to improve memory and other cognitive

functions is currently unknown. Low intensity walking has been shown to improve verbal

memory, executive functions and processing speed (Papp et al. 2014; Scherder et al. 2014) as

well as reduce dementia risk (Fenesi et al. 2016; Groot et al. 2016). Yet, evidence from younger

adults suggests that higher intensity exercises, which are capable of inducing greater changes in

cardiorespiratory fitness, may be needed to improve high-interference memory (Bherer et al.

2013; Déry et al. 2013; Heisz et al. 2017). This may be because high intensity exercise protocols

induced greater increases in BDNF (Afzalpour et al. 2015; Marquez et al. 2015). Although

cardiorespiratory fitness and cognitive performance are positively related in older adults as

evidenced by observation studies (Bherer et al. 2013; Bullock et al. 2018), it is unclear whether

higher intensity exercise causes greater improvements in high-interference memory and general

cognitive functions (Sink et al., 2015; Young et al. 2015).

In addition to being a more potent physiological stimulus than low or moderate intensity

exercise (Gibala and McGee 2008), high intensity exercise is also more time efficient. This has

important implications for uptake by the older adult population. Regardless of one’s age,

ethnicity, sex or health status, the most commonly cited reason for not exercising is a lack of

time (Godin et al. 1994). Thus, an exercise prescription that requires minimal time commitment

may be an effective approach to enhancing activity for this population (Gibala and McGee

2008). Similar to traditional moderate intensity endurance training (i.e., continuous brisk

walking or jogging), high intensity exercise (i.e., at about 90% of maximum heart rate capacity)

induces numerous metabolic adaptations in older adults including increased skeletal muscle

oxidative capacity and endurance performance (Wyckelsma et al. 2017); the added benefit is that

Page 5 of 38



Draft


high intensity exercise sessions are shorter in duration and require fewer sessions to yield

positive results (Gibala and McGee 2008), and thus may be more feasible in the long term.

This community-based study investigated the effect of aerobic exercise intensity on

cognition and BDNF in sedentary but otherwise healthy older adults. Participants were assigned

to one of three intervention groups: a high intensity interval-training group (HIIT), a moderate

intensity continuous training group (MCT), or a stretching control group (CON). High-

interference memory was tested using an adapted mnemonic similarity task (Groot et al. 2016;

Stark et al. 2015; Stark et al. 2013). Executive functions were examined using the Go-Nogo and

Flanker tasks. Peripheral BDNF concentrations were quantified using enzyme-linked

immunosorbent assays (ELISAs). We hypothesized that that high intensity exercise training

would lead to the greatest increases in BDNF and improvements in high-interference memory.

Materials and Methods

Participants

This community-based study took place over a period of approximately 2.5 years from

August 2014 to March 2017. The methodology was originally reported in a master’s thesis

(Kovacevic 2017) and is described in full below. A large effect size (r2 > 0.5) was expected for

the primary outcome based on a previous study using a high-interference memory task (Déry et

al. 2013). The software G*Power was used to determine an adequate sample size of 64

participants needed to detect a significant difference between groups with β = .8 and α = .05. Due

to the rolling nature of the protocol, participants were over-recruited (n = 83) to account for

dropouts.

Participants were recruited through local news outlets and postings. Participants

consisted of sedentary, but otherwise healthy community-dwelling adults over the age of 60

Page 6 of 38



Draft


years. Exclusion criteria included engaging in more than one hour of vigorous physical activity

per week, or a diagnosis of cognitive impairment. The Montreal Cognitive Assessment (MoCA)

(Nasreddine et al. 2005) was also used to assess baseline cognitive abilities; however, no one

was excluded based on their score due difficulty in assigning a reliable and uniform cut-off

across advancing age (Oren et al., 2015). Eligibility was assessed through verbal or written

confirmation via phone or email. Potentially eligible participants completed a stress test with

their physician prior to enrolment to screen for any abnormal response to physical exercise.

Participants with abnormal responses were deemed ineligible. This study received ethics

clearance from the Hamilton Integrated Research Ethics Board. All participants provided written

informed consent prior to experimental procedures and were compensated $40 upon study

completion.

Figure 1 depicts the overall design of the study. It is described in detail in the following

sections.

[Insert Figure 1 here]

Procedure

The experimental procedure consisted of a pre and post intervention assessment that each

took approximately two hours and included one hour of cognitive assessments, a half hour of

physical and fitness assessments, and a half hour of questionnaire-based assessments. Post-

testing was completed within 48 hours of the final intervention exposure.

Participants trained in one of three groups for the duration of the intervention: 1) High

intensity interval training group (HIIT n=21), 2) Moderate-intensity continuous training group

(MCT; n=20), or 3) Stretching exercise control group (CON; n=23). Participants were assigned

to groups by a researcher. Group assignment was done according to blocks stratified by sex to

Page 7 of 38



Draft


ensure equal distribution amongst groups. Due to equipment availability, group sizes were

limited and therefore participants were assigned to the next available group if applicable. Group

assignment took place prior to pre-intervention assessment, but participants did not learn their

placement until assessments were completed, and all cognitive measures were objective to

minimize bias of unblinding (Kahan et al. 2014). The training protocols and the measures used in

the pre and post intervention assessments are described in the subsequent sections.

Intervention Training

The program was designed such that participants in all groups meet three times per week

for 12 weeks for intervention training supervised by a trained research assistant. Participants

were instructed not to engage in additional physical activity for the duration of the study. The

18th session was replaced with a midpoint exercise test resulting in a total target of 35 training

sessions. To maximize retention, participants were accommodated for absences by continuing to

meet with their respective intervention groups to make up missed sessions until they achieved as

close to the target of 35 training sessions as possible (total weeks trained M=13, SD=3; total

sessions completed M=34, SD=4). All participants completed at least half of the training

protocol.

The HIIT and MCT training protocols were adapted from those described in Wisløff et al.

(2007). To isolate the effect of intensity, the protocols were designed to be isocaloric and

consequently their durations differed to match the total training load (Rognmo et al. 2004;

Wisløff et al. 2007). Exercise training was completed on a motor-driven treadmill (Life Fitness

95Ti). The speed and incline of the treadmill were continually adjusted to ensure participants

were training at their target heart rate, determined from the peak heart rate achieved in the fitness

assessment, and target rating of perceived exertion (RPE) according to the Borg 6-to-20 scale

Page 8 of 38



Draft


(Borg 1982). If both target heart rate and target RPE were not achieved simultaneously, heart

rate was the preferred indicator of intensity achieved.

HIIT. Participants warmed up for three minutes at 0% grade and 50-70% of peak heart

rate, then 10 minutes at 5% grade and 60-70% of peak heart rate. Participants then walked four

four-minute intervals at 5% grade and 90-95% of peak heart rate (target RPE=16-18+). These

intervals were separated by three-minute active recovery periods where participants walked at

50-70% of peak heart rate (target RPE=9-11). After the final high-intensity interval, participants

walked a three-minute cool-down at 50-70% of peak heart rate. A final two-minute cool-down

was completed following the pre-programmed protocol on the treadmill. Total exercise time was

43 minutes. Heart rate and RPE were recorded at the end of each interval.

MCT. Participants warmed up for three minutes at 0% grade and 50-70% of peak heart

rate before walking continuously at 70-75% of peak heart rate for 47 minutes (target RPE=12-

14). A final two-minute cool-down was completed following the pre-programmed protocol on

the treadmill, as above. Total exercise time was 52 minutes. Heart rate and RPE were recorded

after the warm-up and every seven minutes for the remainder of the session.

CON. Participants engaged in a series of non-aerobic seated and standing stretches. The

program was specifically designed for older adults and aimed at whole-body stretching. Each

session was 30 minutes long. RPE was recorded after each 30-minute session. These sessions

took place in a large classroom.

Pre and Post Intervention Assessment

Memory. High-interference memory was tested using an adapted Mnemonic Similarity

Task (Groot et al. 2016; Stark et al. 2015; Stark et al. 2013). Participants began with an encoding

phase in which 60 full color images of everyday objects on a white background were presented

Page 9 of 38



Draft


in random order on a computer screen for 2 seconds each. A blank screen preceded each trial for

500 milliseconds. Participants were instructed to classify each item as indoor or outdoor using

the “1” and “2” keys, respectively, on the number pad. Immediately following the encoding

phase, participants engaged in a forced-choice visual recognition phase, in which they classified

items as “Old” (repetitions), “Similar” (lures), or “New” (foils) using the “1”, “2”, or “3” key,

respectively, on the number pad. The recognition phase consisted of 90 trials presented in

random order for an unlimited time until the participant responded: 30 stimuli were repeated

from the encoding phase (repetitions), 30 were similar, but not identical (lures), and 30 were not

previously viewed (foils). Performance on the task, corrected for response bias, was assessed

using the lure discrimination index: the difference between the proportion of correctly identified

lures as “Similar” minus the proportion of incorrectly classified foils as “Similar”

[p(“Similar”|Lure)-p(“Similar”|Foil)] (Stark et al. 2015; Stark et al. 2013).

The data were checked to ensure that participants performed the task correctly.

Classifying foils as “New” is relatively simple, therefore participants who did not classify any

foils as “New” were thought to not have understood the task instructions. One participant’s data

for the MST were discarded for this reason.

Executive functions. Executive functions were tested using two computer tasks: the Go-

Nogo task (Clark et al. 2015), and the Flanker task. In this task, participants were presented with

120 letters in white font on a black background in random order on a computer screen for 500

milliseconds each. They were instructed to respond as quickly as possible by pressing the space

bar when they saw any letter (go trials) except for “A”, “S”, or “F” (nogo trials). The nogo trials

comprised one third of the total number of trials. A blank screen preceded the presentation of a

stimulus for a jittered duration between 500-1000 milliseconds.

Page 10 of 38



Draft


For the Flanker task, a fixation cross was centered on the computer screen for one

second, after which a single set of five arrows appeared for 50 milliseconds. The set was either

congruent (i.e., “<<<<<” or “>>>>>”) or incongruent (i.e., “<<><<” or “>><>>”). There were

90 trials in total and one third were incongruent. Participants were instructed to indicate whether

the central arrow in the set pointed left or right by pressing the “z” or “/” key, respectively. There

were an equal number of left and right target responses.

Performance on both executive functions tasks was assessed using an inverse efficiency

score (IES) (Bruyer and Brysbaert 2011). The IES is calculated as the mean response time of the

correct responses on go trials for the Go-Nogo and on incongruent trials for Flanker divided by

the proportion of trials that were responded to correctly. By including accuracy, this calculation

accounts for the controlled processing required to complete the task in addition to the speed of

response.

Cardiorespiratory fitness. Fitness assessment to predict peak oxygen uptake (VO2 peak)

was performed at pre, mid, and post testing using a modified Bruce protocol on a motor-driven

treadmill (Life Fitness 95Ti). Each stage of the protocol was three minutes long. Participants

walked at a speed of 1.7 mph and 0% grade in the first stage, and 1.7 mph and 5% grade in the

second stage, in accordance with the Bruce protocol modifications made by Willemsen et al.

(2010). The remainder of the test followed the standardized Bruce protocol (Sheffield and

Roitman 1976). A trained research assistant supervised the test and recorded time at exhaustion,

heart rate at each interval and at exhaustion (measured using Polar FT1 heart rate monitors), and

RPE at each interval according to the Borg 6-to-20 scale (Borg 1982). The test was terminated

upon volitional exhaustion or presentation of abnormal symptoms. Predicted VO2 peak was

calculated using Equation 1 below, where the weighting factor was 1 for men and 2 for women

Page 11 of 38



Draft


(Bruce et al. 1973). The duration used in the equation was that achieved on the standardized

Bruce protocol, excluding the six minutes added as a result of the modification (Willemsen et al.

2010). Participants who did not complete protocol beyond the modification were excluded from

fitness analyses (N=3). Since this protocol was a measure of predicted cardiorespiratory fitness,

participants’ results (VO2 peak and peak heart rate) on the modified Bruce protocol were

correlated with performance on the exercise stress test, completed prior to study enrolment, to

ensure that the predicted measures were valid.

Predicted VO2 peak = 6.70–2.82 (weighting factor for sex)+0.056 (duration in seconds) (1)

Participants completed a familiarization at pre-test to get accustomed to walking on the

treadmill and the full modified Bruce protocol was completed at a subsequent visit. During the

familiarization, participants completed 9.5 minutes (the first 3 stages) of the modified Bruce

protocol, unless volitional exhaustion was reached earlier. An additional fitness assessment was

completed at midpoint to reassess peak heart rate and increase the difficulty of exercise training

if a greater peak heart rate was achieved. The first 11 participants in the study completed the

Single Stage Treadmill Walking Test, a submaximal cardiorespiratory fitness test (Ebbeling et al.

1991) instead of the maximal modified Bruce protocol, which was subsequently used for all

participants upon additional ethics clearance for this specific protocol. Participants who

completed the submaximal cardiorespiratory fitness test were excluded from analysis of

cardiorespiratory fitness because cardiorespiratory fitness values obtained from the submaximal

test did not correlate with those obtained from their physician’s stress test (p = .53) and therefore

the values of the submaximal test were considered unreliable.

BDNF. Peripheral blood samples were collected from participants in the morning

following 12 hours of fasting and prior to the cardiorespiratory fitness test described above.

Page 12 of 38



Draft


Serum samples were obtained from whole blood collected in BD Vacutainer SST tubes (BD,

Franklin Lanes, NJ, USA). Upon collection, the tubes were kept at room temperature for 30

minutes to clot and then were centrifuged at 4000 rpm for 10 minutes at 4°C. The supernatant

was then aliquoted into Eppendorf tubes and stored at -80°C until analysis.

Serum BDNF concentrations were quantified using an enzyme-linked immunosorbent

assay (ELISA) according to kit specifications using the human free BDNF Quantikine ELISA kit

(R&D Systems; Minneapolis, MN, USA; cat#SBD00). All samples and standards were run in

duplicate and the optical density of samples was determined after ten minutes of adding Stop

Solution at 450 nm with wavelength correction set to 540 nm. Samples were measured using

BioTek’s Synergy Mx Microplate Reader and Gen5 1.11 Microplate Reader and Imager

Software (BioTek).

Statistical Analyses

Data were checked for outliers and normality, and to ensure appropriate assumptions

were met for each analysis. Outliers were defined as values beyond quartiles 1 (Q1) and 3 (Q3)

with a step of 1.5 times the interquartile range (IQR; i.e., values <Q1–1.5IQR or >Q3+1.5IQR).

Overall, there were very few outliers, which were removed pairwise (cardiorespiratory fitness:

HIIT=1; memory: MCT=2, CON = 1; Flanker: HIIT=3; BDNF: HIIT=1). Change scores

collapsed across group were also independently evaluated for outliers, with few outliers removed

pairwise (memory: CON=1; GoNogo: CON=1; Flanker: HIIT=1, CON=2; BDNF: MCT=3,

CON=4). Normality was assessed by histograms and significance on the Kolmogorov-Smirnov

test. Only the Flanker results were non-normal and reciprocally transformed prior to analysis.

Homogeneity of variance was assessed using Levene’s test for ANCOVAs.

Page 13 of 38



Draft


As a manipulation check, the mean number of sessions attended, and heart rate and RPE

during exercise were calculated to ensure that the intended prescription was achieved.

To confirm that the interventions induced the expected fitness adaptations from pre- to

post-test, a univariate ANCOVA was conducted on post-test predicted VO2 peak values with a

between-subjects factor of group, and with age and pre-test predicted VO2 peak values entered as

covariates. To verify that the modified Bruce protocol used in this study was a valid measure of

both cardiorespiratory fitness and peak heart rate, two two-tailed partial correlations, controlling

for age, were conducted: one between predicted VO2 peak achieved on the modified Bruce

protocol and VO2 peak achieved on the stress test, and one between peak heart rate achieved on

the modified Bruce protocol and peak heart rate achieved on the stress test.

To evaluate how exercise intensity affected cognitive performance, univariate

ANCOVAs were conducted on the post-test scores of each measure using a between-subjects

factor of group. Age and pre-test scores were used as covariates. Post hoc pairwise comparisons

were performed to examine group effects.

Based on prior research, we hypothesized that increases in fitness would lead to

monotonic improvements cognition (Bullock et al. 2018). To examine this, we conducted one-

tailed Spearman’s correlations between VO2 peak change scores and change scores for each

cognitive outcome collapsed across all groups. All change scores were determined by

subtracting pre-test values from post-test values.

Results

Participant flow through the study is presented in Figure 2. A total of 83 interested and

eligible participants attended the pre-testing session. Five participants withdrew before training

and thus 78 were enrolled in the study. Thirteen participants (17%) withdrew during training

Page 14 of 38



Draft


(HIIT=3, MCT=5, CON=5). One participant was excluded from analysis upon becoming

ineligible near study completion because their physician diagnosed them with mild cognitive

impairment. This resulted in 64 participants (61% females; mean age=72 years; age range=60-

88 years) who completed the study and were included in the analysis.

[Insert Figure 2 here]

Baseline demographic data are represented in Table 1. Predicted VO2 peak on the

modified Bruce protocol was significantly correlated with VO2 peak from the stress test (p <

.001). Peak heart rate achieved on the modified Bruce protocol was significantly correlated with

peak heart rate from the stress test (p < .001).

[Insert Table 1 here]

Manipulation check

All analyzed data met the outlined assumptions including normality and homogeneity of

variance. Data from the Flanker task were positively skewed and were reciprocally transformed

to a normal distribution.

Exercise training characteristics are presented in Table 2. Although RPE was lower than

expected for the desired intensity, target heart rate was used as the primary indicator of intensity.

Cardiorespiratory fitness. The interventions induced cardiorespiratory fitness adaptations

from pre- to post-test, such that exercise training led to the greatest increases in predicted VO2

peak [F(2, 43) = 26.45, p < .001, ηp2 = .55]. Pairwise comparisons revealed that both HIIT (29.8

± 4.9) and MCT (30.3 ± 3.8) had greater predicted VO2 peak at post-test than CON (18.7 ± 6.8;

HIIT vs. CON: p < .001; Cohen’s d = 1.9; MCT vs. CON: p < .001, Cohen’s d = 2.1). There

were no differences between HIIT and MCT (p = .83).

[Insert Table 2 here]

Page 15 of 38



Draft


Group-based analyses

Memory. Figure 3 depicts high-interference memory performance across groups. There

was a significant main effect of group on high-interference memory [F(2, 55) = 6.04, p = .004,

ηp2 = .18]. Pairwise comparisons revealed that HIIT induced better memory performance than

MCT (Cohen’s d = 1.2, p = .003) and CON (Cohen’s d = 1.1, p = .004) but MCT and CON were

not significantly different from each other (p = .85).

As depicted in Figure 4, increases in predicted VO2 peak was positively correlated with

improvements in high-interference memory performance with a medium effect size [rs(46) = .27,

p = .03].

[Insert Figure 3 and 4 here]

Executive functions. For the Go-Nogo task, there was a trend towards a main effect of

group [F(2, 59) = 2.54, p = .088, ηp2 = .08]. Both HIIT and MCT had a trend toward inducing

better performance than CON with moderate effect sizes (HIIT vs. CON: Cohen’s d = .7, p =

.056; MCT vs. CON; Cohen’s d = .7, p = .057). There were no differences between HIIT and

MCT (p = .99).

For the Flanker task, there was also trend towards a main effect of group [F(2, 56) = 2.41,

p = .099, ηp2 = .08]. HIIT had a trend toward inducing better performance than both MCT

(Cohen’s d = .7, p = .072) and CON (Cohen’s d = .7, p = .049) with moderate effect sizes. There

was no significant difference between MCT and CON (p = .91).

Increases in predicted VO2 peak tended to correlate with improvements in performance

efficiency on the Go-Nogo task with a medium effect size but not significant [rs(46) = -.23, p =

.06], but there was no correlation with the Flanker task [rs(47) = -.02, p = .44].

Page 16 of 38



Draft


BDNF. Table 3 presents BDNF values across groups. There was no main effect of

group on BDNF [F(2, 54) = .21, p = .81]. Change in BDNF did not correlate with change in

predicted VO2 peak [rs(38) = -.16, p = .32] or change in memory [rs(50) = .04, p = .79].

Likewise, baseline BDNF did not correlate with change in predicted VO2 peak [rs(45) = .14, p =

.34] or change in memory [rs(57) = -.01, p = .93].

Discussion

Higher intensity exercise improved memory in sedentary older adults over a relatively

short 12-week intervention. This is in contrast to more traditional forms of moderate continuous

exercise that had no effect on memory at the group level. Furthermore, improvements in

cardiorespiratory fitness correlated with improvements in memory, suggesting that adaptations in

the utilization of oxygen during exercise may influence brain function in aging.

High-interference memory significantly improved following HIIT but not MCT (Figure

3). This is an important finding because it helps to clarify inconsistencies in the literature with

respect to exercise and memory (Asl et al. 2008; Clark et al. 2015; Creer et al. 2010; Erickson et

al. 2011; Heisz et al. 2015). Specifically, the present study suggests that exercise intensity may

be a critical factor. Previous interventions of similar duration (~12 weeks) but using low-to-

moderate intensity exercise have failed to demonstrate memory improvements in older adults

(Iuliano et al. 2017; Perri and Templer 1985). Although these studies would suggest that memory

improvements may not be possible following such short interventions, we recently demonstrated

that only six weeks of high-intensity exercise training improved memory in young adults (Heisz

et al. 2017). Therefore, the present study provides novel data demonstrating that exercise can

improve memory in older adults over a short intervention, but that the exercise may need to be of

higher intensity.

Page 17 of 38



Draft


At the individual level, we observed a significant correlation between increases in

cardiorespiratory fitness and improvements in memory (Figure 4). Data points for MCT

overlapped with HIIT and so despite robust group level differences favoring HIIT over MCT,

MCT may be of sufficient intensity to enhance memory for some. Interestingly, many

individuals in the CON group had declines in both fitness and memory. This observation

suggests that some exercise, even if not enough to enhance memory, may help mitigate age-

related decline in fitness to prevent memory loss.

In contrast, exercising at either moderate or high intensity had a moderate-sized but

nonsignificant effect on executive functions. Prior interventions of longer duration (i.e., lasting at

least 6 months) that have used moderate intensity exercise and found similar but significant

improvements in executive functions in older adults (Colcombe et al. 2004; Kramer et al. 2001).

Critically, the moderate effect of HIIT and MCT on executive functions was different from the

robust benefit effect of HIIT on high-interference memory and suggests that exercise intensity

may differentially impact these cognitive functions. Indeed, brain processes that subserve

executive functions and memory seem to respond differently to increasing exercise intensity and

this may be related to their sensitivity to physiological stress. For example, the prefrontal cortex

(PFC) is recruited for executive functions. Moderate intensity exercise—a moderate

physiological stressor—evokes a more positive effect on executive functions than more vigorous

exercise (Chang et al. 2012; Labelle et al. 2013). Likewise, activation of the PFC also follows an

inverted-U shaped curve such that a moderate stressor increases glutamatergic transmission to

enhance executive functions (Yuen et al. 2009) whereas a severe stressor impedes this process

(Hains and Arnsten 2008). Although we observed similar effects of both moderate and high

intensity exercise on executive functions, the interval structure of our high intensity exercise

Page 18 of 38



Draft


protocol (i.e., including rest breaks between intense bouts) may have made it overall less severe,

and thus still within the positive range.

BDNF has been identified as a potential mechanism for the effects of exercise on

memory. Unlike the inverted U activation pattern of the PFC, the relationship between exercise

intensity and BDNF seems to be more linear, such that higher intensity exercises elicit greater

BDNF (Ferris et al. 2007; Gold et al. 2003; Saucedo Marquez et al. 2015; Schmidt-Kassow et al.

2012; Vega et al. 2006). Prior work in older adults has demonstrated that high intensity exercise

can increase BDNF to improve memory (De Assis and Almondes 2017); however, that was not

the case here such that we observed significantly group differences in memory but no group

differences in BDNF. Although the results suggest that BDNF may not be mediating theses

cognitive improvements, this is opposite to many reports in the literature (De Assis and

Almondes 2017; Szuhany et al. 2015). Alternatively, it is also possible that the null findings may

be due to the timing of blood sample collection. In previous studies showing exercise-induced

increases in BDNF (Ferris et al. 2007; Schmidt-Kassow et al. 2012), blood samples were

collected within one hour following an exercise bout. Indeed, exercise-induced increases in

BDNF tend to return to baseline within minutes to a few hours following an acute exercise bout

(Vega et al. 2006). Our blood samples were collected 24 to 48 hours after the final exercise

exposure, therefore, any transient changes in BDNF from an acute bout of exercise would have

returned to baseline by the time the samples were collected. Although we were hoping to capture

adaptations in the resting state levels of BDNF from the chronic training program, the results

suggest that the neurotrophic effects of exercise may be closely tied to the acute stimulation from

the exercise bout itself rather than adaptations in the baseline state. Future research is needed to

contrast the impact of exercise training on BDNF levels during the acute phase following a bout

Page 19 of 38



Draft


of exercise as well as at rest. In addition, there may be other biochemical factors, besides BDNF,

that underlie the effect of exercise on memory. One theory based on recent evidence is that HIIT,

being a more intense stimulus than MCT, increases lactate levels to a greater extent. Lactate is an

important factor for promoting neuroplasticity (El Hayek et al. 2019) and may support our

finding of memory improvement following HIIT. Future studies should also include a measure of

lactate to further investigate this potential mechanism.

Overall, this community study provides critical information about the implementation of

exercise programs for memory, but it is not without limitations. The first wave of participants

(n=11) was not included in our assessment of cardiorespiratory fitness because they did not

complete the maximal aerobic fitness test, and thus were treated as missing data. However, the

missing data were distributed across all three groups and so was unlikely to impact the results.

Additionally, participants were classed as sedentary if they reported engaging in less than one

hour of vigorous physical activity per week. While this cut-off is below the 150 minutes of

moderate-to-vigorous aerobic exercise weekly recommended for adults 65 years and older by the

Canadian Society for Exercise Physiology (CSEP 2011), self-reports by participants may not

have most accurately captured true baseline activity levels. We also had broad inclusion criteria

to be more realistic of a community-based program, including people on beta-blockers, which

can interfere with accurate heart rate reading. Although we used both heart rate and RPE as

indicators of intensity, it should be noted that the HIIT group had no participants known to be on

beta-blockers and this may have resulted in a more accurate estimate of heart rate for that group.

With the rapidly growing population of older adults and increasing prevalence of

neurodegenerative disease, it is essential to identify effective interventions for promoting

cognitive function. This community-based study provides much needed evidence regarding the

Page 20 of 38



Draft


effect of exercise intensity on cognitive functions among older adults and helps inform possible

mechanisms for mitigating cognitive decline and potentially dementia risk. Overall the findings

suggest that a relatively short-term program of high-intensity interval exercises may help older

adults improve memory, and thus may represent as an effective strategy for promoting physical

fitness and brain health in aging. Future directions include investigating more direct measures of

memory, such as anterior and posterior hippocampal volume following HIIT versus MCT to

confirm exercise-induced changes in brain physiology. Additionally, it would be important to

investigate the acute time course of BDNF following HIIT versus MCT, as well as other

potential molecular regulators such as lactate.

Acknowledgements

We would like to thank the NeuroFitLab members and the McMaster Physical Activity

Centre of Excellence (PACE) faculty and staff who helped facilitate the project. Most of all, we

are grateful to the participants who dedicated their time to take part in this study.

Funding

This research was supported by the Alzheimer Society of Brant, Haldimand, Norfolk,

Hamilton and Halton, Banting Foundation Discovery Award, and Natural Sciences and

Engineering Research Council of Canada (NSERC) Discovery Grant 296518 to Jennifer J. Heisz.

Ana Kovacevic was supported by a Canada Graduate Scholarship-Master’s from NSERC.

Conflicts of Interest

The authors have no conflicts of interest to report.

Page 21 of 38



Draft


References

Afzalpour, M. E., Chadorneshin, H. T., Foadoddini, M., and Eivari, H. A. 2015. Comparing

interval and continuous exercise training regimens on neurotrophic factors in rat brain.

Physiol. Behav. 147: 78-83. doi: 10.1016/j.physbeh.2015.04.012.

Asl, N. A., Sheikhzade, F., Torchi, M., Roshangar, L., and Khamnei, S. 2008. Long-term regular

exercise promotes memory and learning in young but not in older rats. Pathophysiology,

15(1): 9-12. doi: 10.1016/j.pathophys.2007.10.002.

Bekinschtein, P., Oomen, C. A., Saksida, L. M., and Bussey, T. J. 2011. Effects of environmental

enrichment and voluntary exercise on neurogenesis, learning and memory, and pattern

separation: BDNF as a critical variable? Semin. Cell Dev. Biol. 22(5): 536-542. doi:

10.1016/j.semcdb.2011.07.002.

Bherer, L., Erickson, K. I., and Liu-Ambrose, T. 2013. A review of the effects of physical

activity and exercise on cognitive and brain functions in older adults. J. Aging Res. 2013:

657508. doi: 10.1155/2013/657508.

Borg, G. A. 1982. Psychophysical bases of perceived exertion. Med. Sci. Sports Exerc. 14(5):

377-381. PMID: 7154893

Bruce, R. A., Kusumi, F., and Hosmer, D. 1973. Maximal oxygen intake and nomographic

assessment of functional aerobic impairment in cardiovascular disease. Am. Heart J.

85(4): 546-562. doi: 10.1016/0002-8703(73)90502-4.

Bruyer, R., and Brysbaert, M. 2011. Combining speed and accuracy in cognitive psychology: is

the inverse efficiency score (IES) a better dependent variable than the mean reaction time

(RT) and the percentage of errors (PE)? Psychol. Belg. 51(1): 5-13. doi: 10.5334/pb-51-

1-5.

Page 22 of 38



Draft


Bullock, A. M., Mizzi, A. L., Kovacevic, A., and Heisz, J. J. 2018. The Association of Aging and

Aerobic Fitness With Memory. Front. Aging Neurosci. 10: 63. doi:

10.3389/fnagi.2018.00063.

Canadian Society for Exercise Physiology (CSEP). 2011. Canadian Physical Activity Guidelines

for Older Adults (65 years and older) [online]. Available from

https://csepguidelines.ca/adults-65 [accessed 17 August 2019].

Clark, I. B., Vandermorris, S., and Heisz, J. J. 2015. Physical fitness and sociocognitive

engagement are associated with different aspects of cognition in older adults. J. Am.

Geriatr. Soc. 63(1): 177-179. doi: 10.1111/jgs.13215.

Coelho, F. G. d. M., Vital, T. M., Stein, A. M., Arantes, F. J., Rueda, A. V., Camarini, R., et al.

2014. Acute aerobic exercise increases brain-derived neurotrophic factor levels in elderly

with Alzheimer's disease. J. Alzheimers Dis. 39(2): 401-408. doi: 10.3233/JAD-131073.

Colcombe, S. J., Erickson, K. I., Scalf, P. E., Kim, J. S., Prakash, R., McAuley, E., et al. 2006.

Aerobic exercise training increases brain volume in aging humans. J. Gerontol. A Biol.

Sci. Med. Sci. 61(11): 1166-1170. doi: 10.1093/gerona/61.11.1166.

Colcombe, S. J., Kramer, A. F., Erickson, K. I., Scalf, P., McAuley, E., Cohen, N. J., et al. 2004.

Cardiovascular fitness, cortical plasticity, and aging. Proc. Natl. Acad. Sci. U.S.A.

101(9): 3316-3321. doi: 10.1073/pnas.0400266101.

Cotman, C. W., and Berchtold, N. C. 2002. Exercise: a behavioral intervention to enhance brain

health and plasticity. Trends Neurosci. 25(6): 295-301. doi: 10.1016/s0166-

2236(02)02143-4.

Page 23 of 38



Draft


Creer, D. J., Romberg, C., Saksida, L. M., van Praag, H., and Bussey, T. J. 2010. Running

enhances spatial pattern separation in mice. Proc. Natl. Acad. Sci. U. S. A. 107(5): 2367-

2372. doi: 10.1073/pnas.0911725107.

Curlik, D., and Shors, T. 2013. Training your brain: Do mental and physical (MAP) training

enhance cognition through the process of neurogenesis in the hippocampus?

Neuropharmacology, 64: 506-514. doi: 10.1016/j.neuropharm.2012.07.027.

Daselaar, S. M., Iyengar, V., Davis, S. W., Eklund, K., Hayes, S. M., and Cabeza, R. E. 2013.

Less wiring, more firing: low-performing older adults compensate for impaired white

matter with greater neural activity. Cereb. Cortex, 25(4): 983-990. doi:

10.1093/cercor/bht289.

De Assis, G. G., and Almondes, K. M. D. 2017. Exercise-dependent BDNF as a modulatory

factor for the executive processing at individuals in course of cognitive decline. A

systematic review. Front. Psychol. 8(584): 1-8. doi: 10.3389/fpsyg.2017.00584.

Déry, N., Pilgrim, M., Gibala, M., Gillen, J., Wojtowicz, J. M., MacQueen, G., and Becker, S.

2013. Adult hippocampal neurogenesis reduces memory interference in humans:

opposing effects of aerobic exercise and depression. Front. Neurosci. 7(66): 1-15. doi:

10.3389/fnins.2013.00066.

Ebbeling, C. B., Ward, A., Puleo, E. M., Widrick, J., and Rippe, J. M. 1991. Development of a

single-stage submaximal treadmill walking test. Med. Sci. Sports Exerc. 23(8): 966-973.

PMID: 1956273.

El Hayek, L., Khalifeh, M., Zibara, V., Assaad, R. A., Emmanuel, N., Karnib, N., et al. 2019.

Lactate mediates the effects of exercise on learning and memory through SIRT1-

dependent activation of hippocampal brain-derived neurotrophic factor (BDNF). J.

Page 24 of 38



Draft


Neurosci. 39(13): 2369-2382. doi: 10.1523/JNEUROSCI.1661-18.2019.

Erickson, K. I., Voss, M. W., Prakash, R. S., Basak, C., Szabo, A., Chaddock, L., et al. M. 2011.

Exercise training increases size of hippocampus and improves memory. Proc. Natl. Acad.

Sci. U.S.A. 108(7): 3017-3022. doi: 10.1073/pnas.1015950108.

Fenesi, B., Fang, H., Kovacevic, A., Oremus, M., Raina, P., and Heisz, J. J. 2016. Physical

Exercise Moderates the Relationship of Apolipoprotein E (APOE) Genotype and

Dementia Risk: A Population-Based Study. J. Alzheimers Dis. 56(1): 297-303. doi:

10.3233/JAD-160424.

Ferris, L. T., Williams, J. S., and Shen, C.-L. 2007. The effect of acute exercise on serum brain-

derived neurotrophic factor levels and cognitive function. Med. Sci. Sports Exerc. 39(4):

728-734. doi: 10.1249/mss.0b013e31802f04c7.

Fjell, A. M., McEvoy, L., Holland, D., Dale, A. M., Walhovd, K. B., and Initiative, A. s. D. N.

2014. What is normal in normal aging? Effects of aging, amyloid and Alzheimer's disease

on the cerebral cortex and the hippocampus. Prog. Neurobiol. 117: 20-40. doi:

10.1016/j.pneurobio.2014.02.004.

Fu, Y., Zhang, Y., and Yuan, Q. 2017. Aerobic exercise ameliorates learning and memory

deficits of aging rats induced by D-galactose via promoting SYP and BNDF expression

in hippocampus. In BIO Web Conf. 8: 01020. EDP Sciences. doi:

10.1051/bioconf/20170801020.

Gibala, M. J., and McGee, S. L. 2008. Metabolic adaptations to short-term high-intensity interval

training: a little pain for a lot of gain? Exerc. Sport Sci. Rev. 36(2): 58-63. doi:

10.1097/JES.0b013e318168ec1f.

Page 25 of 38



Draft


Godin, G., Desharnais, R., Valois, P., Lepage, L., Jobin, J., and Bradet, R. 1994. Differences in

perceived barriers to exercise between high and low intenders: observations among

different populations. Am. J. Health Promot. 8(4): 279-285. doi: 10.4278/0890-1171-

8.4.279.

Gold, S. M., Schulz, K.-H., Hartmann, S., Mladek, M., Lang, U. E., Hellweg, R., et al. 2003.

Basal serum levels and reactivity of nerve growth factor and brain-derived neurotrophic

factor to standardized acute exercise in multiple sclerosis and controls. J. Neuroimmunol.

138(1): 99-105. doi: 10.1016/s0165-5728(03)00121-8.

Groot, C., Hooghiemstra, A., Raijmakers, P., van Berckel, B., Scheltens, P., Scherder, E., et al.

2016. The effect of physical activity on cognitive function in patients with dementia: a

meta-analysis of randomized control trials. Ageing Res. Rev. 25: 13-23. doi:

10.1016/j.arr.2015.11.005.

Heisz, J. J., Clark, I. B., Bonin, K., Paolucci, E. M., Michalski, B., Becker, S., and Fahnestock,

M. 2017. The effects of physical exercise and cognitive training on memory and

neurotrophic factors. J. Cogn. Neurosci. 29(11): 1895-1907. doi: 10.1162/jocn_a_01164.

Heisz, J. J., Gould, M., and McIntosh, A. R. 2015. Age-related shift in neural complexity related

to task performance and physical activity. J. Cogn. Neurosci. 27(3): 605-613. doi:

10.1162/jocn_a_00725.

Heisz, J. J., Vandermorris, S., Wu, J., McIntosh, A. R., and Ryan, J. D. 2015. Age Differences in

the Association of Physical Activity, Sociocognitive Engagement, and TV Viewing on

Face Memory. Health Psychol. 34(1): 83-88. doi: 10.1037/hea0000046.

Hillman, C. H., Erickson, K. I., and Kramer, A. F. 2008. Be smart, exercise your heart: exercise

effects on brain and cognition. Nat. Rev. Neurosci. 9(1): 58-65. doi: 10.1038/nrn2298.

Page 26 of 38



Draft


Honea, R., Thomas, G. P., Harsha, A., Anderson, H. S., Donnelly, J. E., Brooks, W. M., and

Burns, J. M. 2009. Cardiorespiratory fitness and preserved medial temporal lobe volume

in Alzheimer's disease. Alzheimer Dis. Assoc. Disord. 23(3): 188-197. doi:

10.1097/WAD.0b013e31819cb8a2.

Iuliano, E., Fiorilli, G., Aquino, G., Di Costanzo, A., Calcagno, G., and di Cagno, A. 2017.

Twelve-Week Exercise Influences Memory Complaint but Not Memory Performance in

Older Adults: A Randomized Controlled Study. J. Aging Phys. Act. 25(4): 612-620. doi:

10.1123/japa.2016-0249.

Kovacevic, A. 2017. Investigating the effect of aerobic exercise on cognitive function in older

adults. M.Sc. thesis, Department of Kinesiology, McMaster University, Hamilton, ON.

Kramer, A. F., Hahn, S., McAuley, E., Cohen, N. J., Banich, M. T., Harrison, C., et al. 2001.

Exercise, aging and cognition: healthy body, healthy mind. In Human factors

interventions for the health care of older adults. Edited by A. D. Fisk and W. Rogers.

Lawrence Erlbaum Associates, Hillsdale, N.J. pp. 91-120.

Marquez, C. M. S., Vanaudenaerde, B., Troosters, T., and Wenderoth, N. 2015. High-intensity

interval training evokes larger serum BDNF levels compared with intense continuous

exercise. J. Appl. Physiol. 119(12): 1363-1373. doi: 10.1152/japplphysiol.00126.2015.

Nasreddine, Z. S., Phillips, N. A., Bédirian, V., Charbonneau, S., Whitehead, V., Collin, I., et al.

2005. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild

cognitive impairment. J. Am. Geriatr. Soc. 53(4): 695-699. doi: 10.1111/j.1532-

5415.2005.53221.x.

Oren, N., Yogev-Seligmann, G., Ash, E., Hendler, T., Giladi, N., and Lerner, Y. 2015. The

Montreal cognitive assessment in cognitively-intact elderly: A case for age-adjusted

Page 27 of 38



Draft


cutoffs. J. Alzheimers Dis. 43(1): 19-22. doi: 10.3233/JAD-140774.

Papp, K. V., Kaplan, R. F., Springate, B., Moscufo, N., Wakefield, D. B., Guttmann, C. R., and

Wolfson, L. 2014. Processing speed in normal aging: Effects of white matter

hyperintensities and hippocampal volume loss. Neuropsychol. Dev. Cogn. B. Aging

Neuropsychol. Cogn. 21(2): 197-213. doi: 10.1080/13825585.2013.795513.

Perri, S., and Templer, D. I. 1985. The effects of an aerobic exercise program on psychological

variables in older adults. Int. J. Aging Hum. Dev. 20(3): 167-172. PMID: 6530295.

Rognmo, Ø., Hetland, E., Helgerud, J., Hoff, J., & Slørdahl, S. A. 2004. High intensity aerobic

interval exercise is superior to moderate intensity exercise for increasing aerobic capacity

in patients with coronary artery disease. Eur. J. Cardiovasc. Prev. Rehabil. 11(3): 216-

222. doi: 10.1097/01.hjr.0000131677.96762.0c.

Saucedo Marquez, C. M., Vanaudenaerde, B., Troosters, T., and Wenderoth, N. 2015. High-

intensity interval training evokes larger serum BDNF levels compared with intense

continuous exercise. J. Appl. Physiol. 119(12): 1363-1373. doi:

10.1152/japplphysiol.00126.2015.

Scherder, E., Scherder, R., Verburgh, L., Königs, M., Blom, M., Kramer, A. F., and Eggermont,

L. 2014. Executive functions of sedentary elderly may benefit from walking: a systematic

review and meta-analysis. Am. J. Geriatr. Psychiatry, 22(8): 782-791. doi:

10.1016/j.jagp.2012.12.026.

Schmidt-Kassow, M., Schädle, S., Otterbein, S., Thiel, C., Doehring, A., Lötsch, J., and Kaiser,

J. 2012. Kinetics of serum brain-derived neurotrophic factor following low-intensity

versus high-intensity exercise in men and women. Neuroreport, 23(15): 889-893. doi:

10.1097/WNR.0b013e32835946ca.

Page 28 of 38



Draft


Scullin, M. K., and Bliwise, D. L. 2015. Sleep, cognition, and normal aging: integrating a half

century of multidisciplinary research. Perspect. Psychol. Sci. 10(1): 97-137. doi:

10.1177/1745691614556680.

Sheffield, L. T., and Roitman, D. 1976. Stress testing methodology. Prog. Cardiovasc. Dis.

19(1): 33-49. doi: 10.1016/0033-0620(76)90007-4.

Stark, S. M., Stevenson, R., Wu, C., Rutledge, S., and Stark, C. E. 2015. Stability of age-related

deficits in the mnemonic similarity task across task variations. Behav. Neurosci. 129(3):

257-268. doi: 10.1037/bne0000055.

Stark, S. M., Yassa, M. A., Lacy, J. W., and Stark, C. E. 2013. A task to assess behavioral

pattern separation (BPS) in humans: Data from healthy aging and mild cognitive

impairment. Neuropsychologia, 51(12): 2442-2449. doi:

10.1016/j.neuropsychologia.2012.12.014.

Vega, S. R., Strüder, H. K., Wahrmann, B. V., Schmidt, A., Bloch, W., and Hollmann, W. 2006.

Acute BDNF and cortisol response to low intensity exercise and following ramp

incremental exercise to exhaustion in humans. Brain Res. 1121(1): 59-65. doi:

10.1016/j.brainres.2006.08.105.

Voss, M. W., Prakash, R. S., Erickson, K. I., Basak, C., Chaddock, L., Kim, J. S., et al. 2010.

Plasticity of brain networks in a randomized intervention trial of exercise training in older

adults. Front. Aging Neurosci. 2(32): 1-17. doi: 10.3389/fnagi.2010.00032.

Willemsen, S., Hartog, J. W. L., Hummel, Y. M., Posma, J. L., van Wijk, L. M., van Veldhuisen,

D. J., and Voors, A. A. 2010. Effects of alagebrium, an advanced glycation end-product

breaker, in patients with chronic heart failure: study design and baseline characteristics of

the BENEFICIAL trial. Eur. J. Heart Fail. 12(3): 294-300. doi: 10.1093/eurjhf/hfp207.

Page 29 of 38



Draft


Wisløff, U., Støylen, A., Loennechen, J. P., Bruvold, M., Rognmo, Ø., Haram, P. M., et al. 2007.

Superior cardiovascular effect of aerobic interval training versus moderate continuous

training in heart failure patients: a randomized study. J. Am. Heart Assoc. 115(24): 3086-

3094. doi: 10.1161/CIRCULATIONAHA.106.675041.

Wyckelsma, V. L., Levinger, I., McKenna, M. J., Formosa, L. E., Ryan, M. T., Petersen, A. C.,

et al. 2017. Preservation of skeletal muscle mitochondrial content in older adults:

relationship between mitochondria, fibre type and high-intensity exercise training. J.

Physiol. 595(11): 3345-3359. doi: 10.1113/JP273950.

Yassa, M. A., and Stark, C. E. L. 2011. Pattern separation in the hippocampus. Trends Neurosci.

34(10): 515-525. doi: 10.1016/j.tins.2011.06.006.

Zheng, G., Xia, R., Zhou, W., Tao, J., and Chen, L. 2016. Aerobic exercise ameliorates cognitive

function in older adults with mild cognitive impairment: a systematic review and meta-

analysis of randomised controlled trials. Br. J. Sports Med. 50(23): 1443-1450. doi:

10.1136/bjsports-2015-095699.

Page 30 of 38



Draft


Tables

Table 1. Demographic characteristics at baseline across groups

Note. Data represented are mean and standard deviation (SD) or number of participants. MoCA: Montreal Cognitive Assessment (Nasreddine et al. 2005); bpm: beats per minute. *means and SD based on fewer number of participants with available data (Predicted VO2 peak, HIIT=17, MCT=17, Control=18; Stress test VO2 peak, HIIT=15, MCT=12, CON=15; Peak heart rate, HIIT=17, MCT=17, CON=18; Stress test peak heart rate, HIIT=20, MCT=20, CON=22).† One-way ANOVA‡ Pearson Chi-Square

HIIT (N=21) MCT (N=20) CON (N=23)p

Age (years) 72.4 (4.4) 72.0 (6.2) 71.5 (6.6) .87†

Sex (female) 14 10 15 .48‡

MoCA (max score = 30) 25 (3) 26 (3) 26 (2) .53†

Body Mass Index (kg/m2) 27 (4) 28 (4) 30 (6) .18†

Blood pressure (mmHg)

Systolic

Diastolic

138.7 (17.8)

73.1 (9.3)

139.7 (23.8)

71.3 (11.1)

140.2 (19.4)

69.0 (8.6)

.97†

.37†

β-blockers

No

Yes

Unknown

16

0

5

15

3

2

15

3

5

.35‡

Predicted VO2 peak (ml/(kg•min))* 25.0 (6.2) 24.9 (5.5) 19.2 (6.7) .009†

Stress test VO2 peak (ml/(kg•min))* 21.1 (3.0) 20.4 (4.1) 18.1 (3.7) .07†

Peak heart rate (bpm)* 142.7 (12.9) 138.5 (14.5) 134.0 (24.2) .37†

Stress test peak heart rate (bpm)* 142.9 (17.0) 137.7 (19.2) 129.7 (23.7) .11†

Page 31 of 38



Draft

32

Table 2. Intervention training characteristics across groups

Outcome HIIT (N=21) MCT (N=20) CON (N=23) p

Training sessions attended 33.8 (3.6) 32.5 (5.1) 34.9 (1.7) .11†

Number of weeks attended 12.9 (2.6) 12.5 (3.2) 13.9 (2.0) .21†

Session heart rate

Mean heart rate

Mean percentage of peak

125.1 (9.2)

87.4 (3.6)

104.8 (11.9)

74.6 (5.2)

N/A

N/A

<.001*

<.001*

Mean session RPE (6-20) 12.9 (1.9) 9.3 (1.6) 8.3 (1.7) <.001†a

Note. Data represented are mean and standard deviation (SD). N/A: not applicable; RPE: Rated perceived exertion.† One-way ANOVA*Independent samples T Testa HIIT>MCT (p<0.001), HIIT>CON (p<0.001), MCT>CON (p=0.075)

Page 32 of 38



Draft

33

Table 3. Mean (SD) for cardiorespiratory fitness, cognition and BDNF at pre- and post-test across groupsHIIT MCT CON

Pre Post Pre Post Pre Post

Outcomen Mean (SD) n Mean (SD) n Mean (SD) n Mean (SD) n Mean (SD) n Mean (SD)

Cardiorespiratory Fitness

Predicted VO2 peak (ml/(kg•min))

17 25.0 (6.2) 15 29.8 (4.9)b 17 24.9 (5.5) 17 30.3 (3.8)b 18 19.2 (6.7) 16 18.7 (6.8)

Cognition

High-interference memory

21 0.26

(0.15)

21 0.34 (0.14)a,b 19 0.24

(0.24)

17 0.17 (0.14) 22 0.12

(0.12)

22 0.17

(0.17)Go IES (ms) 21 494 (66) 21 488 (51)d 19 502 (65) 20 485 (64)d 23 522 (72) 23 529 (71)

Go reaction time (ms) 21 479 (56) 21 484 (51) 19 482 (57) 20 462 (51)b 23 507 (63) 23 518 (65)Go accuracy (%)

21 97

(3)21 99

(1)19 96

(5)20 96

(5)23 97

(3)23 98

(2)Flanker IES (ms)e 21 1863 (2140) 18 818 (174)c,d 20 1505 (1431) 20 1274 (920) 23 1368 (1064) 23 1315 (1054)

Flanker reaction time (ms)

21 905 (526) 18 686 (133) 20 954 (510) 20 865 (315) 23 887 (429) 23 903 (610)

Flanker accuracy (%) 21 71 (24) 18 85 (8) 20 77 (19) 20 78 (17) 23 78 (23) 23 78 (21)

BDNF (ng/mL) 19 29.5 (6.8)

19 27.6(6.7)

19 29.1(7.9)

19 26.0(7.8)

22 24.6 (10.4)

21 25.0 (10.3)

Note. Data are represented with mean and standard deviation (SD); RT = reaction time; IES = inverse efficiency score

a significantly different from MCT (p<0.05)b significantly different from CON (p<0.05)c trending difference from MCT (p<0.1)d trending difference from CON (p<0.1)e analysis conducted on reciprocally transformed data, means reflect raw values with outliers from analysis removed

Page 33 of 38



Draft

34

Figure Captions

Figure 1. Schematic of study design.

Figure 2. Flow of participants through study protocol.

Figure 3. Mean high-interference memory performance for high-intensity interval training (HIIT), moderate continuous training (MCT) and control (CON) at post-test. Covariates appearing in the model were evaluated at the following values: Age = 72, pre-test = 19. **p < .01. Bars represent standard error.

Figure 4. Association between change in cardiorespiratory fitness and memory performance across high-intensity interval training (HIIT), moderate continuous training (MCT) and control (CON): rs(46) = .27, p = .03.

Page 34 of 38



Draft

Pre-Intervention Assessment

Intervention training (12 weeks)

HIIT (43min) MCT (52min) CON (30min)Warm-up Warm-up

Interval (4min) x4Target HR: 90-95% peak

Target RPE: 16-18+

Recovery (3min) x4Target HR: 50-70% peak

RPE: 9-11

Continuous walking (47min)

Target HR: 70-75% peakTarget RPE: 12-14

Cool-down Cool-down

Whole-body stretching (30min)

Post-Intervention Assessment

Figure 1. Schematic of study design.

Page 35 of 38



Draft

Figure 2. Flow of participants through study protocol.

Allocated to High-intensity group (n=28) Received allocated

intervention (n=24) Did not receive

allocated intervention (n=4)- Ineligible (n=1)- Health limitations

(n=3)

Allocation

Analysis (N=64)

Follow-Up

Enrolled (n=83)

Allocated to Control group (n=28) Received allocated

intervention (n=28)

Allocated to Moderate-intensity group (n=27) Received allocated

intervention (n=26) Did not receive

allocated intervention due to health limitations (n=1)

Discontinued intervention (n=5)

Personal health (n=1) Injury unrelated to

intervention (n=1) Dissatisfied with group

assignment (n=2) Reason not specified

(n=1)


Family reasons (n=1) Discomfort from

exercise (n=2) Injury unrelated to

intervention (n=1) Reason not specified

(n=1)


Personal health (n=1) Reason not specified

(n=2)

Analysed (n=21) Analysed (n=20) Excluded from analysis

due to diagnosis of cognitive impairment before study completion (n=1)

Analysed (n=23)

Enrollment

Page 36 of 38



Draft

Figure 3. Mean high-interference memory performance for high-intensity interval training (HIIT), moderate continuous training (MCT) and control (CON) at post-test. Covariates appearing in the model were evaluated

at the following values: Age = 72, pre-test = 19. **p < .01. Bars represent standard error.

38x54mm (300 x 300 DPI)

Page 37 of 38



Draft

Figure 4. Association between change in cardiorespiratory fitness and memory performance across high-intensity interval training (HIIT), moderate continuous training (MCT) and control (CON): rs(46) = .27, p =

.03.

177x186mm (300 x 300 DPI)

Page 38 of 38



Draft · Draft The effects of aerobic exercise intensity on memory in older adults Journal: Applied...

Documents

Transcript of Draft · Draft The effects of aerobic exercise intensity on memory in older adults Journal: Applied...