ORIGINAL RESEARCH

published: 12 March 2021

doi: 10.3389/fragi.2021.632545

A Novel Program to Improve Cognitive

Function in Individuals With Dementia
Using Transcranial Alternating Current
Stimulation (tACS) and Tutored
Cognitive Exercises
Zahra Moussavi *, Kazushige Kimura, Lonnie Kehler, Cristina de Oliveira Francisco and
Brian Lithgow

University of Manitoba, Biomedical Engineering, Winnipeg, MB, Canada

The effects of cognitive exercises on the healthy aging population is controversial.

Transcranial alternating current stimulation (tACS) is considered a promising tool for
modulating brain oscillation. Research is lacking on its long-lasting cognitive/
therapeutic effect. This is the first pilot study to explore the effect of a regimen of
cognitive exercises with and without tACS on older adults with dementia. The study
Edited by: groups were 28 individuals (age 56–83 years) enrolled into two groups: Exr Group, who
Morten Scheibye-Knudsen, received cognitive exercises only and the Exr + tACS Group who received tACS at 40 Hz
University of Copenhagen, Denmark
simultaneously with cognitive exercises for a period of 4 consecutive weeks, 5 days/week,
Reviewed by:
Sebastian J. Hofer, two 30 min-sessions/day; all the training sessions were tutored. The cognitive exercises
University of Graz, Austria were applied using the MindTriggers app. They were assessed at pre and post intervention
Yujun Hou,
National Institutes of Health (NIH),
and also one month after the end of trial (follow-up) with an independent assessment
United States (WMS-IV) as the primary outcome measure. The results show significant cognitive
*Correspondence: improvement at post-intervention in both groups, while the Exr + tACS protocol lead
Zahra Moussavi to superior cognitive improvement at follow-up session. The most important outcomes of
[email protected]
this study are: 1) The tutored repeated practice of the MindTriggers app exercises does
Specialty section: significantly improve the cognitive functions of older adults with dementia and that that
This article was submitted to improvement lasts for at least one month after the end of the intervention, and 2) The
Interventions in Aging,
a section of the journal
application of tACS increases the positive effects of cognitive exercises with the positive
Frontiers in Aging effect lasting an even longer period of time than exercises alone; in other words we
Received: 23 November 2020 speculate that it may lead to long-term potentiation.
Accepted: 08 February 2021
Published: 12 March 2021 Keywords: cognitive exercises, mind triggers app, transcranial alternating current stimulation (tACS), dementia,
tutoring, alzheimer’s, gamma band stimulation
Citation:
Moussavi Z, Kimura K, Kehler L,
de Oliveira Francisco C and Lithgow B INTRODUCTION
(2021) A Novel Program to Improve
Cognitive Function in Individuals With
Due to advances in healthcare and improvements in life style, nowadays, people live longer.
Dementia Using Transcranial
Alternating Current Stimulation (tACS)
However, as longevity increases, cognitive abilities such as executive function, memory,
and Tutored Cognitive Exercises. reasoning, and processing speed deteriorate (Smith et al., 2009). While it can be part of normal
Front. Aging 2:632545. aging (Boudreaus, 2010), a decline in cognitive function or associative and spatial memories can be
doi: 10.3389/fragi.2021.632545 considered a precursor to dementia, in particular Alzheimer’s (Didic et al., 2011). Currently there is

Frontiers in Aging | www.frontiersin.org 1 March 2021 | Volume 2 | Article 632545

Moussavi et al. Dementia: tACS and Cognitive Exercises

no cure or "vaccine" for dementia, however, there is hope to delay stimulation parameter choice (i.e., frequency, intensity, and
the onset of the disease and/or slow its progression by keeping a anatomical location) has been demonstrated in vitro and in
brain wise-healthy life style. This hope is based on the vivo experiments (Thut et al., 2017). The interest in tACS has
neuroplasticity of the brain achievable even in old age as well increased significantly in the past decade. Its potential to
as the individuals’ cognitive reserve (Stern, 2012). Constructing a influence neural oscillations in a frequency- and phase-specific
“reserve capacity” has been suggested to help seniors maintain manner offers the possibility to demonstrate causal relations
cognitive function (Baltes and Baltes, 1990). This theory was later between oscillations and behavior (Neuling et al., 2013; Bland
supported by neuroimaging studies that showed increased and Sale, 2019). Furthermore, the largely untapped therapeutic
contralateral hemispheric activity in right frontal regions for potential of tACS has inspired a number of researchers to study
both working memory (Reuter-Lorenz et al., 2000) and its applicability as a treatment option for numerous neurological
episodic memory (Cabeza et al., 2002). and psychological disorders (Kensinger et al., 2003). However,
In recent years, there have been many programs and studies there has been little investigation into the potential effects of tACS
using different types of brain exercises claiming that they can on older adults (Antonenko et al., 2016) and no investigation on
enhance the cognitive function of older adults if they are used individuals with dementia. Moreover, while much of the research
frequently and regularly (Willis et al., 2006; Garcia-Camupzano on tACS involves testing participants’ cognitive performance
et al., 2013; Park and Bischof, 2013). However, improvements during or immediately following stimulation sessions (Jaušovec
were observed mostly in the trained tasks (near-effect) with little and Jaušovec, 2014; Meiron and Lavidor, 2014; Hoy et al., 2015;
or no transfer of cognitive functional enhancement to similar but Antonenko et al., 2016), a cumulative (over a few weeks) tACS
untrained tasks (far-effect) (Karbach and Verhaeghen, 2014). program has not been explored. In addition, the majority of the
Several review papers on meta-analysis of computerized above-mentioned studies have applied tACS not only for a short
training programs also found no evidence for far-effect for any duration but also without any other cognitive learning; only one
working memory exercises on older adults (Karbach and study (Antonenko et al., 2016) applied tACS during a language
Verhaeghen, 2014; Lampit et al., 2014). Another review study learning paradigm. While, there is no study investigating the
found no conclusive evidence of the effect cognitive exercises had repeated training with tACS, from studies using transcranial
on individuals with dementia and also questioned the amount of direct current stimulation (tDCS), it is known that the
training required to receive any cognitive benefit from the stimulation alone without any cognitive learning does not lead
exercises (Christie et al., 2017). A very recent study to any long-lasting improvement (Huo et al., 2018).
(Stojanoski et al., 2018), running a spatial training task and In this pilot study, for the first time, we explore the effect of a
another similar testing task on 72 healthy participants regimen of cognitive exercises using the MindTriggers app
(between 20 and 62 years), in particular, showed evidence (available on iPad AppStore) with and without tACS (called
against either a near-effect or far-effect gain from cognitive Exr + tACS and Exr groups, respectively) on older adults with
function training. dementia. The number of participants in our study is small but
It is important to note that almost all studies on the effect of comparable with all other similar studies on the same topic. This
brain exercises on the cognitive function of adults and older is part of an on-going clinical trial (Moussavi, 2018) that repeats
adults had the participants doing the exercises on their own with the training program every 4 months on individuals with mild to
no interaction with a trainer. In contrast to the above-mentioned moderate dementia. This paper presents data and outcomes from
study (Stojanoski et al., 2018), our previous and recent studies the first round of tACS treatment program, whilst the repeated
(Garcia-Camupzano et al., 2013; Garcia-Campuizano, 2013; treatment data is being collected. We hypothesize that both
Kehler et al., 2020) and also our recent study on the effect of treatment groups’ cognitive status will improve at post-
brain exercises on older adults with memory problems intervention and that improvement will persist at least one
(unpublished) have shown significant cognitive improvement month. In addition, we hypothesize that the tACS + Exr group
in individuals with mild dementia when they received the will show more improvement than the Exr group. We
brain exercises within a regimen comprising a tutored learning investigated our hypotheses using the independent outcome
environment delivered on a daily basis; this improvement measures at post-intervention and follow-up sessions
transferred to their daily life beyond the practiced exercises compared to baseline.
(the far-effect).
In addition to, or independent of brain exercises, some
researchers have explored application of transcranial METHOD
alternating current stimulation (tACS) on cognition. tACS is a
relatively inexpensive, easy to administer and safe tool for non- Participants’ Data
invasive brain stimulation (Antal and Paulus, 2013); it utilizes the Data from 28 individuals with various degrees of dementia were
external application of oscillating electrical currents to influence adopted from the pool of data of an ongoing clinical trial studying
cortical activity. The tACS-generated oscillating electric field has the effects of cognitive exercises with or without tACS on the
been demonstrated to both modulate and entrain the ongoing aging population (Moussavi, 2018). Participants of the above
network oscillations in a frequency-specific manner (Antal and clinical trial were enrolled into tACS + Exr and Exr groups
Paulus, 2013). The feasibility of using tACS for manipulating the predominantly based on their tolerability to tACS and their
phase, rhythm and power of neural oscillations via appropriate preference; there was no randomization in group assignment.

Frontiers in Aging | www.frontiersin.org 2 March 2021 | Volume 2 | Article 632545

Moussavi et al. Dementia: tACS and Cognitive Exercises

TABLE 1 | Participants Demographic information

Characteristics tACS + Exr Exr

N (M/F) 19 (13/6) 9 (7/2)

Age (years) (mean ± SE) 73.1 ± 1.8 69.5 ± 3
MoCA (mean ± SE) 18.4 ± 1.2 16.4 ± 1.2
Diagnosis •8 AD (2 moderate, 6 mild) • 5 AD
•3 LBD • 1 VD
•1 FTD • 2 MCI
•1 PCA
•1 VD
•4 MCI
•1 neurodegenerative dementia

TABLE 2 | WMS Memory index scores calculation.

Memory Index Sub-test score components Maximum score

Auditory Memory Logical Memory I & II, and Verbal Paired Associates I & II 142
Visual Memory Visual Reproduction I & II 86
Visual Working Memory Symbol Span 50
Immediate Memory Logical Memory I, Verbal Paired Associates I, and Visual Reproduction I 136
Delayed Memory Logical Memory II, Verbal Paired Associates II, and Visual Reproduction II 92
Recognition Memory Logical Memory II Recognition, Verbal Paired Associates II Recognition, Verbal Paired Associates II Word Recall, and Visual 80
Reproduction II Recognition

Participants who could not focus on exercises while receiving • Have no history of epileptic seizure or stroke.
tACS or found tACS painful, were enrolled into Exr group.
Therefore, our two groups of intervention are not matched in Since the COVID19 pandemic, we have additionally offered
size or gender in this pilot study. However, as the clinical trial this program Online but data of those participants have also been
continues, we will generate balanced groups in the future. excluded from the analysis in this paper.
All volunteers are assessed by Montreal Cognitive Assessment Data of 28 participants who met the above criteria, were
(MoCA) (Nasreddine et al., 2005). The MoCA is a 30-point test to selected for analysis in this study, out of which 3 felt the
assess cognitive function. This test takes approximately 20 min to tACS as painful and 6 felt the tACS as tolerable but
administer in this population, and screens for mild cognitive uncomfortable when doing the exercises at baseline
impairment and dementia. A score of 26 or higher is considered session; thus, 19 were enrolled in tACS + Exr and 9 in Exr
as normal cognition. This test is used only for screening. groups. Table 1 shows demographic information of the
The study was been approved by the Biomedical Research participants and their diagnosis, whose data were analyzed
Ethics Board of the University of Manitoba and all volunteers in this study.
signed a consent form prior to enrollment into any of the study
groups. Cognitive Exercises
The selection of the data for this study was based on the We used the exercises within the MindTriggers app
following criteria: (Moussavi, 2020) delivered on an iPad. The exercises in
this app were conceptually designed by the first author
• Age >60 years old based on neuroplasticity of the brain and the most
• A MOCA score of 7–25 impaired cognitive functions in individuals with dementia,
• Be diagnosed with either mild cognitive impairment (MCI) particularly Alzheimer’s. The app includes 7 different serious
or mild/moderate stage of a dementia subtype by a games with different difficulty levels. The focus of the games
physician. are on strengthening associative and spatial cognitive skills
• Except for the dementia diagnosis, have no known diagnosis that deteriorate even in normal brain aging (Boundreaus,
of major depression, bipolar disorder, schizophrenia, mood 2010). A significant decline in these two types of memory are
disorder, Parkinson’s disease, Huntington disease, ALS, MS, symptoms of MCI, dementia and in particular Alzheimer’s.
and/or any other neurological disorder. Associative memory is defined as the creation of new links
• Have a Reading/writing/comprehension fluency in English between different items which were not related previously,
• Able to be enrolled in the 4-weeks, 5 days/week program of and the later retrieval of these new associations (Buschkuehl
the clinical trial. and Jaeggi, 2010). Spatial memory includes the memory of

Frontiers in Aging | www.frontiersin.org 3 March 2021 | Volume 2 | Article 632545

Moussavi et al. Dementia: tACS and Cognitive Exercises

FIGURE 1 | The VR building used for spatial orientation assessment.

FIGURE 2 | WMS raw scores of all participants of the two groups at baseline, post and follow-up sessions. Vertical axis is the WMS raw scores and horizontal axis is
the participants’ number with no particular order. Note that three participants in tACS + Exr and 1 in Exr group missed the follow-up session assessment due to
pandemic.

geographical layout of an environment, and is important for Transcranial Alternating Current

orientation and navigation. Studies show recalling associative Stimulation (tACS)
links between different items in older adults requires We used the Soterix Medication tACS Stimulator (Model: 2001)
significant effort and energy (Buschkuehl and Jaeggi, on the tACS group. The stimulation was applied with a sinusoidal
2010). These types of memory or cognition require a waveform at a frequency of 40Hz and a current amplitude of
healthy connectivity between associated parts of the brain. −0.75 mA to +0.75 mA (1.5 mA p-p). The active electrode was
For example, to recognize a familiar object and being able to placed over the left dorsolateral prefrontal cortex (DLPFC) and
name it; the right parietal lobe is activated to identify the object, the reference electrode was placed on the contralateral
it then communicates with the temporal lobe on the left side of supraorbital area. Stimulation was applied simultaneously with
the brain to name it. In this process, of course the hippocampus, the brain exercises per each session of treatment on participants
which is central for long-term memory formation, and the in tACS + Exr group.
prefrontal cortex which is involved in decision making, also
become active and involved. Thus, an effective exercise for
maintaining a healthy brain or improving the cognitive status Training/Treatment Protocol
of dementia population must target exercising these areas of the The treatment protocol is to deliver the cognitive exercises of
brain and their connectivity; that has been the premise of the MindTriggers app (Moussavi, 2020) on an iPad with a trained
designed games. Appendix A details the MindTriggers exercises tutor of our team on a daily basis for 4 weeks consecutively
and their scoring. (excluding weekends, 20 sessions in total). Each daily treatment

Frontiers in Aging | www.frontiersin.org 4 March 2021 | Volume 2 | Article 632545

Moussavi et al. Dementia: tACS and Cognitive Exercises

TABLE 3 | Participants’ performance data, Mean ± SE window (target appears on each front/back/left/right sides of the
Measure/Groups Baseline Post Follow-up building in either left or right corners of either the second or third
floor). The building is then rotated clockwise so that the
WMS-IV Total tACS 95.2 ± 9.7 114 ± 12.3 125.2 ± 16.9 participant can see every side of the building from outside.
Non-tACS 75.3 ± 9.2 97.3 ± 13.4 101.5 ± 15.7
After a full rotation, the participant is asked to verbally say
MADRS tACS 3.7 ± 1.1 2.7 ± 1 1.5 ± 0.6
Non-tACS 3.1 ± 0.7 1.9 ± 0.6 2 ± 1.1 where the target window with the X mark is. This is called
Spatial % tACS 49.3 ± 8.1 67.9 ± 6 64.7 ± 8.1 “target localization” stage, in which a target was placed on
Non-tACS 47.8 ± 15.5 66 ± 7.4 71.4 ± 13 each left, right and back of the building twice in each corner
of the wall. This stage of the test was run 6 times to cover the 6
different options as the target. If a participant identified the target
correctly, s/he would receive 3 scores: 2 for identifying the target
includes two 30 min duration cognitive exercises. The tACS + Exr on a correct wall/side of the building and 1 for identifying the
group receive tACS simultaneously during the cognitive exercises. target on the correct corner of the wall. We do not assign a score
Each participant has a dedicated tutor throughout the period of for floor. As this test is run 6 times, the maximum correct score
the study. is 18.
Montgomery Asberg Depression Rating Scale (MADRS) -
Assessments The MADRS (Montgomery and Asberg, 1979) is a ten-item
The following assessments were run at baseline, post-intervention rating scale for rating levels of depression. Each item is rated
on first day of week 5 and a month after the end of the trial on a 7-point scale from 0 to 6, where 0 indicates absence of the
(follow-up). Although the cognitive exercises of MindTriggers symptom and 6 indicates extreme presence of the symptom. The
have an analytic scoring system (Appendix A), we did not use time frame for the scale is the previous four weeks. MADRS was
those as metrics for measuring improvements for two main used to test for the presence of depression as a confounding
reasons: 1) the training sessions were tutored; thus, the scores variable.
do not reflect the true ability of the participants, and 2) the games’
scores, at best, measure the near-effect of the training program, Statistical Analysis
which includes the practice effects of doing the same task We hypothesis that the daily training with MindTriggers’
repeatedly. We are interested in investigating the far-effect of cognitive exercises will improve the cognitive abilities of
the treatment program that can be assessed only by independent study participants in both tACS + Exr and Exr groups when
but relevant assessments as described below. assessed by the outcome measures pre to post-intervention.
Primary Outcome Measure: Wechsler Memory Scale (WMS Moreover, we hypothesize the plausible positive effect of the
IV) Older Adult Battery - This assessment (Cullum and cognitive exercises will last at least a month. Thus, we should
Larrabee, 2010) has six major memory indexes: Auditory not see a significant decline at Follow-up assessment with
Memory, Visual Memory, Visual Working Memory, respect to immediate post-intervention. In addition, we
Immediate Memory, Delayed Memory, and Recognition hypothesize the group who receive tACS will be cognitively
Memory. We calculated a total WMS score, which has a
maximum score of 358, by summing all the sub-test scores,
while we calculated memory index scores by summing related
sub-test sections as shown in Table 2; note that memory indexes
have overlap in terms of the sub-tests’ scores.
Secondary Outcome Measure: Egocentric Spatial
Orientation - This is a virtual reality (VR), focused on route
finding and mapping in a cubic 3-story building with no
landmark or cue; this test developed and evaluated in our lab
for detection of spatial cognitive impairment (Byagowi et al.,
2014; Ranjbar-Pooya et al., 2017). This test has two stages of
assessment: 1) target localization from outside of the building,
and 2) target finding inside the building. Most of those with MCI
or dementia are not able to perform the target finding part as they
find it very challenging. In this study, we only used the scores of
the localization stage of the test using a laptop screen (not in an
immersive mode).
In brief, the VR building in this test is a cubic symmetric FIGURE 3 | The average of primary outcome measure (WMS-IV)
landmark-less environment, with 4 rooms (with windows to (mean ± SE) among participants of the two groups of tACS + Exr (n  19) and
outside) on each side of the building, 2 rooms/windows on Exr (n  9) at baseline, post-intervention and follow-up. Both groups showed
each of the second and third floors in left and right corners of significant improvement at post-intervention respect to baseline; this
improvement was still significant for tACS + Exr group but not for Exr group;
the wall (Figure 1). One of the windows in each trial of the eight
see Table 4 for statistical analysis details.
trials is pseudo randomly marked with a big X mark as the target

Frontiers in Aging | www.frontiersin.org 5 March 2021 | Volume 2 | Article 632545

Moussavi et al. Dementia: tACS and Cognitive Exercises

FIGURE 4 | The average changes (mean ± SE) of the WMS-IV raw scores among participants of the two groups of tACS + Exr (n  19) and Exr (n  9) at post-
intervention and follow-up respect to baseline. The changes between the two groups at follow-up session is evident but it was not statistically significant (p  0.086); see
Table 4 for details on statistical analysis details.

significantly better than the Exr group at follow-up assessment showed negligible (only 1 or 2 scores of WMS-IV) decline in their
session. scores. Figure 2 shows the WMS raw scores of all participants of
Both groups’ data were checked for normality. If the outcome the two groups at three assessment sessions. It is important to
measures passed the normality, we used repeated measure note that individuals with dementia decline gradually; thus, any
Multivariate Analysis of Variance (MANOVA) and its post- program to help to avoid decline can be considered an
hoc analysis and to investigate the above three hypotheses. If improvement. The rate of decline, however, is neither linear
data did not pass the normality test, we used the equivalent non- nor the same among different dementia type. Therefore, it
parametric tests, the Wilcoxon Signed-Ranks test. In all statistical would be challenging to give an expected decline rate for our
tests, p  0.05 was considered as significant. study participants as their cognitive impairment levels and causes
varied widely.
None of the participants were suffering from depression at
RESULTS baseline and their depression scores remained approximately
level throughout the sessions with small improvements at
Three participants in tACS group and one in non-tACS group did post-intervention. Table 3 shows the average and standard
not have follow-up assessment scores due to the COVID19 error of the WMS-IV total score, MADRS and Spatial scores
pandemic but all had the post-intervention assessment. at baseline, post-intervention and follow-up sessions.
All participants of both groups except 3 in the tACS + Exr and Repeated measure of MANOVA showed a significant
1 in the Exr groups, did improve in overall cognitive scores improvement from baseline to post-intervention in both tACS
significantly when assessed by WMS-IV. Those 4 participants, + Exr and Exr groups (Figure 3), confirming the first hypothesis.

TABLE 4 | Summary of the statistical tests in WMS scores. Adjusted p-values are bolded, and the p-value with *, ** and *** indicates significance at p < 0.05, p < 0.01 and
p < 0.001, respectively. The bold values imply statistical significance.

Analysis Statistic p-value

Group * Time (interaction term) F (1.43, 30.05)  0.824 0.412

Group (tACS + Exr and Exr) F (1, 21)  0.068 0.797
Time (baseline, post-intervention and follow-up) F (1.43, 30.05)  13.997 0.0002***
tACS + Exr group from baseline to post-intervention t (18)  −5.33 0.0001***
tACS + Exr group from baseline to follow-up t (15)  −4.06 0.003**
tACS + Exr group from post-intervention to follow-up t (15)  −1.83 0.262
Exr group from baseline to post-intervention t (8)  −3.26 0.034*
Exr group from baseline to follow-up t (6)  −2.07 0.251
Exr group from post-intervention to follow-up t (6)  0.519 1.0
Change from baseline to post-intervention between the two groups t (12.66)  −0.39603 0.699
Change from baseline to follow-up between the two groups t (13.404)  0.84264 0.414
Change from post-intervention to follow-up between the two groups t (20.87)  1.8015 0.086

Frontiers in Aging | www.frontiersin.org 6 March 2021 | Volume 2 | Article 632545

Moussavi et al. Dementia: tACS and Cognitive Exercises

TABLE 5 | Summary of the statistical tests in MADRS scores.

Statistic p-value

Group * Time (interaction term) F (2, 44)  0.181 0.835

Group (tACS + Exr and Exr) F (1, 22)  0.079 0.781
Time (baseline, post-intervention and follow-up) F (2, 44)  2.206 0.122
Change from baseline to post-intervention between the two groups t (21.775)  0.054164 0.957
Change from baseline to follow-up between the two groups t (14.794)  −0.53776 0.599
Change from post-intervention to follow-up between the two groups t (14.61)  −0.055052 0.957

As can be seen, both groups kept their improved state at the As one may claim the improvements might be due to
follow-up session, but interestingly the tACS + Exr group showed decreased depression among participants, we also investigated
continued improvement confirming the second and third the MADRS scores among the two groups. The Mixed ANOVA
hypotheses. Figure 4 shows the changes (improvements) from on MADRS scores did not reveal any significant interaction
baseline to post and follow-up sessions for each group, in which a between the groups and time. Similarly, the change in the
significant difference between the two groups is evident at follow- MADRS scores between the two groups did not differ
up although it was not statistically significant. Details of statistical significantly; see Table 5 for details.
analysis are as follow. The same analyses were applied for the secondary outcome
We analysed the outcome measures data using multivariate measure, the spatial assessment scores. As Figure 5 shows,
analysis of repeated measures ANOVA with “Time” as within- participants of both groups showed similar improvements on
subjects factor with 3 levels (baseline, post-intervention and average after the intervention. However, both groups’
follow-up) and “Group” as a Between-Subjects factors. Table 4 improvements at post-intervention and follow-up sessions
shows the significant effects and interactions. As for the primary were also the same. The Mixed ANOVA did not reveal a
outcome measure (WMS-IV), mixed ANOVA revealed a significant interaction between group and time (F (2, 38) 
significant main effect of time (F (1.43, 30.05)  13.997, p  0.249, p  0.781; effect size  0.004) and the main effect of the
0.0002; effect size  0.039). Thus, we performed post-hoc group (F (1, 19)  0.107, p  0.747; the effect size is 0.004), but
analysis with Bonferroni correction to investigate which revealed a significant main effect of time (F (2, 38)  3.948, p 
group showed significant change in their WMS score at 0.028; effect size  0.059). Similar to the analyses of WMS scores,
which time of the assessment (i.e., baseline, post-intervention a post-hoc analysis was designed with Bonferroni correction to
and follow-up). The post-hoc analysis revealed that both groups investigate which group showed significant change in their
showed significant improvements from baseline to post- spatial score at which time of assessment (i.e., baseline, post-
intervention. At the follow-up assessment, the tACS + Exr intervention and follow-up). Despite the significant main effect
group showed better retention of their improved status and on time, none of the pairwise comparisons revealed a significant
even improved further at the follow-up session (a month after change in either of the groups. Aside from the scores, the
the end of intervention). changes in the scores between the assessments were
To investigate the last hypothesis, the changes in the WMS investigated. Similar to the analysis of WMS scores, the
scores between the groups were compared using Welch two tailed difference between the two groups were compared in terms
t-test. The tACS + Exr group showed greater improvement than of the changes in the spatial scores; however, unlike the primary
the Exr Group as expected although not statistically significant outcome measure, no significant changes in spatial scores were
(p  0.08). See Table 4 for details. found; see Table 6 for details.

TABLE 6 | Summary of the statistical tests in spatial scores. Adjusted p-values are bolded, and the p-value with * means significance at p < 0.05. The bold values imply
statistical significance.

Statistic p-value

Groups * Time (interaction term) F (2, 38)  0.249 0.781

Groups (tACS + Exr and Exr) F (1, 19)  0.107 0.747
Time (baseline, post-intervention and follow-up) F (2, 38)  3.948 0.028*
tACS + Exr Group from baseline to post-intervention t (17)  −2.42 0.082
tACS + Exr Group from baseline to follow-up t (13)  −1.23 0.726
tACS + Exr Group from post-intervention to follow-up t (13)  1.02 0.978
Exr Group from baseline to post-intervention t (7)  −1.44 0.582
Exr Group from baseline to follow-up t (6)  −1.53 0.534
Exr Group from post-intervention to follow-up t (6)  −0.148 1.0
Change from baseline to post-intervention between the two groups t (9.924)  −0.77787 0.455
Change from baseline to follow-up between the two groups t (7.3245)  −1.5821 0.156
Change from post-intervention to follow-up between the two groups t (10.214)  −1.5317 0.156

Frontiers in Aging | www.frontiersin.org 7 March 2021 | Volume 2 | Article 632545

Moussavi et al. Dementia: tACS and Cognitive Exercises

FIGURE 5 | The average of secondary outcome measure (Spatial Orientation scores) (mean ± SE) among the two groups of tACS + Exr (n  19) and Exr (n  9) at
baseline, post-intervention and follow-up.

DISCUSSION also may reduce intracerebral Tau protein burden (Dhaynaut

et al., 2020); all these studies have had small sample sizes
In this study, we used MindTriggers app (Moussavi, 2020) that similar to our study. In this study, we observed that both
includes a series of seven cognitive exercises (games) targeting groups (Exr and Exr + tACS) improved significantly and
strengthening the executive function of the brain, its left-right similarly post-intervention; however, the Exr + tACS
hemisphere connectivity and spatial orientation. We tested the continued to improve further during the month after the
efficacy of an intense cognitive exercises program (4 weeks, end of the intervention. The change between the two
5 days/week) on 28 individuals with dementia at mild and groups at follow-up did not reach the significance level,
moderate stages in two groups: A) tACS + Exr Group: which might be due to small sample size. Nevertheless, the
receiving tACS simultaneously with cognitive exercises, and improving trend from post to follow-up in Exr + tACS group
B) Exr Group: receiving only cognitive exercises; both groups implies the application of tACS might have helped with long-
received the cognitive exercises with a trainer at our labs. The term potentiation via enhanced modulation of neural circuitry
results overall showed significant improvement in the cognitive to improve cognition.
status of the participants that lasted at least for a month after On the other hand, application of tACS simultaneously
the end of trial. Analysis of results at post-intervention respect with brain exercises does not seem to have the same effect on
to baseline confirms the first hypothesis that the cognitive spatial cognition. As it is shown in Figure 4, both groups’
exercises program has the main effect in improving the participants have improved their spatial cognition similarly at
cognitive status of the participants as both groups improved both post-intervention and follow-up sessions although
similarly. On the other hand, those who did receive tACS neither changes were found statistically significant between
simultaneously with cognitive exercises improved much the groups. This is an interesting outcome although due the
more (p  0.043) than those without tACS at follow up limited sample size we have to be cautious in our
assessment. This interesting result imply the added positive interpretation. If it is verified in larger sample size, it might
effect of tACS when applied simultaneously with cognitive be due to the location of tACS application. We applied the
exercises in repeated sessions. tACS active electrode on the left DLPFC and its reference on
We applied tACS at 40 Hz in this study. Gamma band contralateral supraorbital area; thus, mostly increasing the
waves (>30 Hz) are known to be the fastest brain waves that excitability of the prefrontal cortex. Spatial egocentric
are generated when the brain is learning a new concept and/or mapping, which was the type of spatial cognition exercised
solving a problem. While research has not shown conclusively and tested in our study, involves several different regions of the
as to whether the dementia patients have a specific deficit in brain but most importantly the posterior parietal cortex
Gamma band activities (McDermott et al., 2018). it has been (Herweg and Kahana, 2018); that was not where we
shown that stimulating the Gamma band activities can lead to stimulated with tACS in this study. Moreover, one should
an enhanced cognitive status in dementia patients (Silva et al., consider that 6 out of the 7 games of MindTriggers app is
2015), improved working memory (Hoy et al., 2015), focused on working memory including associative memory
increased endogenous attention (Hopfinger et al., 2017) and and only one game is dedicated to spatial memory. Likewise,

Frontiers in Aging | www.frontiersin.org 8 March 2021 | Volume 2 | Article 632545

Moussavi et al. Dementia: tACS and Cognitive Exercises

the amount of time that a participant practiced the spatial exercising body and muscles, which normally results in better
memory was only 1/7 of the total time spent on cognitive outcomes.
exercises. It is possible that with more practice on that game,
the results would become significant. Study Limitations
Aside from the more positive effect of tACS at follow-up, the In this study, we did not have a matched size control group to
results of this study is congruent with the outcomes of our measure the practice effect of the assessments. The main reason
previous studies (Garcia-Camupzano et al., 2013; Garcia- was the inability to recruit people in our target population willing
Campuizano, 2013) showing significant improvement in to be only assessed and not receive the training. However, we had
elderly with memory problems after going through tutored 20 one MCI and one early-stage Alzheimer’s who were enrolled into
sessions of brain exercises. It should be noted that all of our the study but could not attend the daily sessions of the training
assessments have been independent of but relevant to the trained due to living out of town; though, they did come in for the
tasks; in fact, we measured the far-effect of our training protocol. assessments. Those two participants’ data show no practice effect
As mentioned in the Introduction, a recent study of the assessment; in fact, they showed a small decline in their
(Stojanoski et al., 2018) claims there is no general gain (no scores at the subsequent assessments at 4 weeks and 8 weeks after
near or far-effect) of practicing brain exercises repeatedly in baseline. In addition, in our previous study (Garcia-Camupzano
healthy individuals. Our studies’ results and in particular this et al., 2013), we had tested the practice effect of WMS-III in an
study’s outcomes argues against that claim. It is worth age-and-gender-matched group using, which was found
comparing the similarities and differences of this study and negligible among older adults who had some memory issues
that in (Stojanoski et al., 2018) although the participants of that but were still considered to be cognitively healthy. Thus, in an
study were healthy individuals (age 20–62) and our MCI/dementia population of this study, the positive
participants were older adults (age 56–83) with dementia. improvements observed most likely was not due to any
Given that if a program is effective for individuals with practice effect of the outcome measures.
dementia it is also expected to be effective for healthy We also acknowledge the limited sample size of this study,
individuals. As there are no other similar studies to compare similar to all previous studies in this subject area that use cognitive
with our study, we compared the cognitive training programs exercises and/or tACS. Recruitment of dementia patients for such
in our study with that in (Stojanoski et al., 2018). For such study with a demanding protocol is challenging. Nevertheless, we
comparison, we only compared the outcomes of EXR group of hope in the near future we increase our sample size.
this study as the other study did not apply tACS. Aside from the limited sample size (similar to all other similar
The protocol of the study in (Stojanoski et al., 2018) was previous studies), another limitation of this study is that we did not
1 month, 5 days/week (20 sessions) training although they have a sham (placebo) group for tACS. Also, the participants’
accepted if a participant attended only 16 sessions of training. enrollment was not randomized between the two groups of the
Similarly, they tested the participants pre- and post-interventions study. The main reason for lack of placebo group was that this
by independent assessment (but similar in concept) to investigate study was funded by donations. For that reason, we felt morally
the far-effect of the training. However, aside the obvious differences responsible to give the participants the real treatment given the limited
such as age group (young adults vs. older adults in our study) and funds. We hope in the future, with a major funding, we can recruit
number of participants, there are several key differences between enough participants to have a large randomized double-blind study.
our study and that in ( Stojanoski et al., 2018) as follows.
The study in (Stojanoski et al., 2018)used only one game (we
had seven different games) called “Token Search” which involves CONCLUSION
working and spatial memory. This game, although proven to be
effective as for a measure of spatial and working memory, can This paper, for the first time, presents the outcomes of our pilot study
become a boring game if played every day for a period of one on repeated sessions of cognitive exercises with and without tACS in a
month. Thus, it is not surprising that out of the 76 individuals regimen incorporating a tutored learning of older adults with various
recruited for the study in (Stojanoski et al., 2018) and who were levels of dementia. The most important outcomes of this study are
paid upon completing the study, only 47 of them completed the twofold: 1) The tutored repeated practice of the MindTriggers app
study. In addition, the participants of the study in ( Stojanoski exercises does significantly improve the cognitive functions of older
et al., 2018) were all young people with very little motivation adults with dementia and that improvement lasts for at least one
other than being paid to participate in the study. On the other month after the end of the intervention, and 2) The application of
hand, among our participants, who were older adults with various tACS improves the positive effects of cognitive exercises with the
degrees of dementia, there were a great desire to improve. positive effect lasting a longer period of time; in other words we
Most importantly, the participants in (Stojanoski et al., 2018) speculate that it may lead to long-term potentiation.
did the test by themselves as they were young healthy adults,
while our participants all had a tutor/instructor sitting with
them during the session and instructing them to choose a game DATA AVAILABILITY STATEMENT
and for how long, and help when necessary. This is a key factor
behind the significant improvement of our participants post- The raw data supporting the conclusions of this article will be
treatment. It is in a way similar to hiring a personal trainer for made available by the authors, without undue reservation.

Frontiers in Aging | www.frontiersin.org 9 March 2021 | Volume 2 | Article 632545

Moussavi et al. Dementia: tACS and Cognitive Exercises

ETHICS STATEMENT BL contributed to paper writing, reviewing and discussion. The

author KK also contributed by assisting to run the program.
The studies involving human participants were reviewed and
approved by the Biomedical Research Ethics Board of the
University of Manitoba. The patients/participants provided FUNDING
their written informed consent to participate in this study.
This study has been supported by Manitoba and British Colombia
family donations including the generous private donation from
AUTHOR CONTRIBUTIONS Puchniak family (Winnipeg, MB). The first author (ZM) owns the
MindTriggers app. The app’s revenue is directed toward the same
ZM was the principle investigator, analyzed the data and wrote donation budget for similar research. None of the participants of
the paper. KK contributed to the statistical analysis. LK assisted in this study paid for the app as they practiced the app with a tutor in
running the assessments. CF assisted in coordinating the study. our labs.

versus exogenous attention. Cogn. Neurosci. 8 (2), 102–111. doi:10.1080/

REFERENCES 17588928.2016.1194261
Hoy, K., Bailey, N., Arnold, S., Windsor, K., John, J., Daskalakis, Z. J., et al. (2015).
Antal, A., and Paulus, W. (2013). Transcranial alternating current stimulation The effect of γ-tACS on working memory performance in healthy controls.
(tACS). Front. Hum. Neurosci. 7, 317. doi:10.3389/fnhum.2013.00317 Brain Cogn. 101, 51. doi:10.1016/j.bandc.2015.11.002
Antonenko, D., Faxel, M., Grittner, U., Lavidor, M., and Flöel, A. (2016). Effects of Huo, L., Zheng, Z., Li, J., Wan, W., Cui, X., Chen, S., et al. (2018). Long-term
transcranial alternating current stimulation on cognitive functions in healthy transcranial direct current stimulation does not improve executive function in
young and older adults. Neural plasticity 2016, 4274127. doi:10.1155/2016/ healthy older adults. Front. Aging Neurosci. 10, 298. doi:10.3389/fnagi.2018.
4274127 00298
Baltes, P., and Baltes, M. (1990). “Psychological perspectives on successful aging: Jaušovec, N., and Jaušovec, K. (2014). Increasing working memory capacity with
the model of selective optimization with compensation,” in Successful aging: theta transcranial alternating current stimulation (tACS). Biol. Psychol. 96,
Perspectives from the behavioral Sciences (European network on Longitudinal 42–47. doi:10.1016/j.biopsycho.2013.11.006
studies on individual Development. Cambridge, England: Cambridge University Karbach, J., and Verhaeghen, P. (2014). Making working memory work: a meta-
Press, 1–34. analysis of executive-control and working memory training in older adults.
Bland, N. S., and Sale, M. V. (2019). Current challenges: the ups and downs of Psychol. Sci. 25 (11), 2027. doi:10.1177/0956797614548725
tACS. Exp. Brain Res. 237, 3071. doi:10.1007/s00221-019-05666-0 Kehler, L., Francisco, C. O., Uehara, M. A., Moussavi, Z., et al. (2020). “The
Boudreaus, E. O. (2010). Individual differences and memory aging Concerns of older effect of transcranial alternating current stimulation(tACS) on cognitive
adults. Baton Rouge, LA: Louisiana State University. functionin older adults with dementia,” in 2020 42nd Annual International
Buschkuehl, M., and Jaeggi, S. M. (2010). Improving intelligence: a literature Conference of the IEEE Engineering in Medicine and Biology Society
review. Swiss Med. Wkly 140 (19), 266–272. doi:10.4414/smw.2010.12852 (EMBC), Montreal, QC, Canada, July 20–24, 2020. doi:10.1109/
Byagowi, A., Mohaddes, D., and Moussavi, Z. (2014). Design and application of a EMBC44109.2020.9175903
novel virtual reality navigational technology (VRNChair). J. Exp. Neurosci. 8, 7. Kensinger, E. A., Shearer, D. K., Locascio, J. J., Growdon, J. H., and Corkin, S.
doi:10.4137/JEN.S13448 (2003). Working memory in mild Alzheimer’s disease and early Parkinson’s
Cabeza, R., Anderson, N. D., Locantore, J. K., and McIntosh, A. R. (2002). Aging disease. Neuropsychology 17 (2), 230. doi:10.1037/0894-4105.17.2.230
gracefully: compensatory brain activity in high-performing older adults. Lampit, A., Hallock, H., and Valenzuela, M. (2014). Computerized cognitive
Neuroimage 17, 1394. doi:10.1006/nimg.2002.1280 training in cognitively healthy older adults: a systematic review and meta-
Christie, G. J., Hamilton, T., Manor, B. D., Farb, N. A. S., Farzan, F., Sixsmith, A., analysis of effect modifiers. Plos Med. 11, e1001756. doi:10.1371/journal.pmed.
et al. (2017). Do lifestyle activities protect against cognitive decline in aging? A 1001756
review. Front. Aging Neurosci. 9, 381. doi:10.3389/fnagi.2017.00381 McDermott, B., Porter, E., Hughes, D., McGinley, B., Lang, M., O’Halloran, M.,
Cullum, C. M., and Larrabee, G. J. (2010). “Chapter 6 - WAIS-IV use in et al. (2018). Gamma band neural stimulation in humans and the promise of a
neuropsychological assessment,” in WAIS-IV Clinical Use and new modality to prevent and treat Alzheimer’s disease. J. Alzheimer’s Dis. 65 (2),
Interpretation, Editor L. G. Weiss, San Diego: Academic Press, 167–187. 363–392. doi:10.3233/jad-180391
Dhaynaut, M., Cappon, D., Paciorek, R., Macone, J., Connor, A., Guehl, N., et al. Meiron, O., and Lavidor, M. (2014). Prefrontal oscillatory stimulation modulates
(2020). Effects of modulating gamma oscillations via 40Hz transcranial access to cognitive control references in retrospective metacognitive
alternating current stimulation (tACS) on Tau PET imaging in mild to commentary. Clin. Neurophysiol. 125 (1), 77–82. doi:10.1016/j.clinph.2013.
moderate Alzheimer’s Disease. J. Nucl. Med 61, 340. 06.013
Didic, M., Barbeau, E. J., Felician, O., Tramoni, E., Guedj, E., Poncet, M., et al. Montgomery, S. A., and Asberg, M. (1979). A new depression scale designed to be
(2011). Which memory system is impaired first in Alzheimer’s disease? sensitive to change. Br. J. Psychiatry 134, 382. doi:10.1192/bjp.134.4.382
J. Alzheimers Dis. 27, 11–22. doi:10.3233/JAD-2011-110557 Moussavi, Z. (2018). Clinical trial: brain fitness APP for cognitive enhancement.,
Garcia-Campuizano, M. (2013). “Investigating the effect of associative memory NIH: US National Library of Medicine. Available at: https://clinicaltrials.gov/.
exercises in aging population,” in Electrical and computer Engineering Moussavi, Z. (2020). MindTriggers. Available at: https://apps.apple.com/us/app/
Winnipeg, MB, Canada: The University of Manitoba. mind-triggers/id1508008769.
Garcia-Camupzano, M. T., Virues-Ortega, J., Smith, S., and Moussavi, Z. (2013). Nasreddine, Z. S., Phillips, N. A., Bédirian, V., Charbonneau, S., Whitehead, V.,
Effect of cognitive training targeting associative memory in the elderly: a small Collin, I., et al. (2005). The montreal cognitive assessment, MoCA: a brief
randomized trial and A longitudinal evaluation. J. Am. Geriatr. Soc. 61, screening tool for mild cognitive impairment. J. Am. Geriatr. Soc. 53 (4), 695.
2252–2254. doi:10.1111/jgs.12574 doi:10.1111/j.1532-5415.2005.53221.x
Herweg, N. A., and Kahana, M. J. (2018). Spatial representations in the human Neuling, T., Rach, S., and Herrmann, C. S. (2013). Orchestrating neuronal
brain. Front. Hum. Neurosci. 12, 297. doi:10.3389/fnhum.2018.00297 networks: sustained after-effects of transcranial alternating current
Hopfinger, J. B., Parsons, J., and Fröhlich, F. (2017). Differential effects of 10-Hz stimulation depend upon brain states. Front. Hum. Neurosci. 7, 161. doi:10.
and 40-Hz transcranial alternating current stimulation (tACS) on endogenous 3389/fnhum.2013.00161

Frontiers in Aging | www.frontiersin.org 10 March 2021 | Volume 2 | Article 632545

Moussavi et al. Dementia: tACS and Cognitive Exercises

Park, D. C., and Bischof, G. N. (2013). The aging mind: neuroplasticity in response Thut, G., Bergmann, T. O., Fröhlich, F., Soekadar, S. R., Brittain, J. S., Valero-
to cognitive training. Dialogues Clin. Neurosci. 15 (1), 109. doi:10.31887/DCNS. Cabré, A., et al. (2017). Guiding transcranial brain stimulation by EEG/
2013.15.1/dpark MEG to interact with ongoing brain activity and associated functions: a
Ranjbar-Pooya, O., Byagowi, A., Kelly, D. M., and Moussavi, Z. (2017). Introducing position paper. Clin. Neurophysiol. 128 (5), 843. doi:10.1016/j.clinph.2017.
a new age-and-cognition-sensitive measurement for assessing spatial 01.003
orientation using a landmark-less virtual reality navigational task. Q. J Exp. Willis, S., Tennstedt, S. L., Marsiske, M., Ball, K., Elias, J., Koepke, K. M., et al.
Psychol. 70 (7), 1406–1419. doi:10.1080/17470218.2016.1187181 (2006). Long-term effects of cognitive training on everyday functional
Reuter-Lorenz, P., Jonides, J., Smith, E. E., Hartley, A., Miller, A., Marshuetz, C., et al. outcomes in older adults. JAMA 296 (23), 2805. doi:10.1001/jama.296.23.
(2000). Age differences in the frontal lateralization of verbal and spatial working 2805
memory revealed by PET. J. Cogn. Neurosci. 12, 174. doi:10.1162/089892900561814
Silva, V. F., Ribeiro, A. P., Andréa Dos Santos, V., Nardi, A. E., Spear King, A. L., Conflict of Interest: ZM is the developer and owner of the MindTriggers app that
and Calomeni, M. R. (2015). Stimulation by light and sound: therapeutics has been used in this study. The apps small revenue is directed to her research
effects in humans. Systematic review. Clin. Pract. Epidemiol. Ment. Health 11, budget for similar studies.
150–154. doi:10.2174/1745017901511010150
Smith, G. E., Housen, P., Yaffe, K., Ruff, R., Kennison, R. F., Mahncke, H. W., et al. The remaining authors declare that the research was conducted in the absence of
(2009). A cognitive training program based on principles of brain plasticity: any commercial or financial relationships that could be construed as a potential
results from the improvement in memory with plasticity-based adaptive conflict of interest.
cognitive training (IMPACT) study. J. Am. Geriatr. Soc. 57 (4), 594–603.
doi:10.1111/j.1532-5415.2008.02167.x Copyright © 2021 Moussavi, Kimura, Kehler, de Oliveira Francisco and Lithgow.
Stern, Y. (2012). Cognitive reserve in ageing and Alzheimer’s disease. Lancet This is an open-access article distributed under the terms of the Creative Commons
Neurol. 11, 1006. doi:10.1016/S1474-4422(12)70191-6 Attribution License (CC BY). The use, distribution or reproduction in other forums is
Stojanoski, B., Lyons, K. M., Pearce, A. A. A., and Owen, A. M. (2018). Targeted permitted, provided the original author(s) and the copyright owner(s) are credited
training: converging evidence against the transferable benefits of online brain and that the original publication in this journal is cited, in accordance with accepted
training on cognitive function. Neuropsychologia 117, 541–550. doi:10.1016/j. academic practice. No use, distribution or reproduction is permitted which does not
neuropsychologia.2018.07.013 comply with these terms.

Frontiers in Aging | www.frontiersin.org 11 March 2021 | Volume 2 | Article 632545