Enhancing learning experiences: EEG-based passive BCI system adapts learning speed to cognitive load in real-time, with motivation as catalyst

Abstract

Computer-based learning has gained popularity in recent years, providing learners greater flexibility and freedom. However, these learning environments do not consider the learner’s mental state in real-time, resulting in less optimized learning experiences. This research aimed to explore the effect on the learning experience of a novel EEG-based Brain-Computer Interface (BCI) that adjusts the speed of information presentation in real-time during a learning task according to the learner’s cognitive load. We also explored how motivation moderated these effects. In accordance with three experimental groups (non-adaptive, adaptive, and adaptive with motivation), participants performed a calibration task (n-back), followed by a memory-based learning task concerning astrological constellations. Learning gains were assessed based on performance on the learning task. Self-perceived mental workload, cognitive absorption and satisfaction were assessed using a post-test questionnaire. Between-group analyses using Mann–Whitney tests suggested that combining BCI and motivational factors led to more significant learning gains and an improved learning experience. No significant difference existed between the BCI without motivational factor and regular non-adaptive interface for overall learning gains, self-perceived mental workload, and cognitive absorption. However, participants who undertook the experiment with an imposed learning pace reported higher overall satisfaction with their learning experience and a higher level of temporal stress. Our findings suggest BCI’s potential applicability and feasibility in improving memorization-based learning experiences. Further work should seek to optimize the BCI adaptive index and explore generalizability to other learning contexts.

Full text

Frontiers in Human Neuroscience 01 frontiersin.org
Enhancing learning experiences:
EEG-based passive BCI system
adapts learning speed to
cognitive load in real-time, with
motivation as catalyst
NoémieBeauchemin
1*, PatrickCharland
2, AlexanderKarran
1*,
JaredBoasen
1,3, BellaTadson
1, SylvainSénécal
1 and
Pierre-MajoriqueLéger
1*
1 Tech3Lab, HEC Montréal, Information Technology Department, Montreal, QC, Canada, 2 Didactics
Department, Université du Québec à Montréal, Montreal, QC, Canada, 3 Faculty of Health Sciences,
Hokkaido University, Sapporo, Japan
Computer-based learning has gained popularity in recent years, providing learners
greater flexibility and freedom. However, these learning environments do not consider
the learner’s mental state in real-time, resulting in less optimized learning experiences.
This research aimed to explore the eect on the learning experience of a novel
EEG-based Brain-Computer Interface (BCI) that adjusts the speed of information
presentation in real-time during a learning task according to the learner’s cognitive
load. Wealso explored how motivation moderated these eects. In accordance with
three experimental groups (non-adaptive, adaptive, and adaptive with motivation),
participants performed a calibration task (n-back), followed by a memory-based
learning task concerning astrological constellations. Learning gains were assessed
based on performance on the learning task. Self-perceived mental workload,
cognitive absorption and satisfaction were assessed using a post-test questionnaire.
Between-group analyses using Mann–Whitney tests suggested that combining BCI
and motivational factors led to more significant learning gains and an improved learning
experience. No significant dierence existed between the BCI without motivational
factor and regular non-adaptive interface for overall learning gains, self-perceived
mental workload, and cognitive absorption. However, participants who undertook
the experiment with an imposed learning pace reported higher overall satisfaction
with their learning experience and a higher level of temporal stress. Our findings
suggest BCI’s potential applicability and feasibility in improving memorization-based
learning experiences. Further work should seek to optimize the BCI adaptive index
and explore generalizability to other learning contexts.
KEYWORDS
brain-computer interface, passive BCI, adaptive interface, EEG, neuroadaptive,
learning, computer-based learning, cognitive load
1 Introduction
Computer-Based Learning (CBL) is an educational approach that uses computer soware
to deliver, assist, and enhance the learning processes (Grizioti and Kynigos, 2020). e CBL
environment learners use in their learning can take multiple forms, such as programs,
applications, tools, and platforms (Grizioti and Kynigos, 2020). CBL provides students with
OPEN ACCESS
EDITED BY
Zachary Freudenburg,
University Medical Center Utrecht,
Netherlands
REVIEWED BY
Vacius Jusas,
Kaunas University of Technology, Lithuania
Wonjun Ko,
Sungshin Women’s University, Republic of
Korea
*CORRESPONDENCE
Noémie Beauchemin
noemie.beauchemin@hec.ca
Alexander Karran
alexander-john.karran@hec.ca
Pierre-Majorique Léger
pml@hec.ca
RECEIVED 12 April 2024
ACCEPTED 26 September 2024
PUBLISHED 07 October 2024
CITATION
Beauchemin N, Charland P, Karran A,
Boasen J, Tadson B, Sénécal S and Léger P-M
(2024) Enhancing learning experiences:
EEG-based passive BCI system adapts
learning speed to cognitive load in real-time,
with motivation as catalyst.
Front. Hum. Neurosci. 18:1416683.
doi: 10.3389/fnhum.2024.1416683
COPYRIGHT
© 2024 Beauchemin, Charland, Karran,
Boasen, Tadson, Sénécal and Léger. This is an
open-access article distributed under the
terms of the Creative Commons Attribution
License (CC BY). The use, distribution or
reproduction in other forums is permitted,
provided the original author(s) and the
copyright owner(s) are credited and that the
original publication in this journal is cited, in
accordance with accepted academic
practice. No use, distribution or reproduction
is permitted which does not comply with
these terms.
TYPE Original Research
PUBLISHED 07 October 2024
DOI 10.3389/fnhum.2024.1416683

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 02 frontiersin.org
instant feedback, individualized learning paths and greater exibility,
all of which can increase student engagement and comprehension
(Grizioti and Kynigos, 2020; Mertens etal., 2022; Van der Kleij etal.,
2015). As a result, CBL is increasingly used in educational programs
as an important complement to conventional classroom teaching or
as a stand-alone pedagogical method (Grizioti and Kynigos, 2020).
However, oering access to CBL does not guarantee a successful
learning experience. For example, online courses allow many more
students to enroll because the number of physical seats available in the
classroom does not limit their capacity. Moreover, their accessibility
makes it possible to take the course at any time, from anywhere in the
world. Because of their greater capacity and the diversity of students
enrolled in these online courses, the vast majority of online courses
have been developed using the classic “one size ts all” approach, with
little to no consideration of individual dierences and learning
abilities (Tekin etal., 2015; Wang and Lehman, 2021). In addition, the
distance between the teacher and the students in CBL makes the
assessment of learning needs and abilities even more dicult (Tek in
et al., 2015). As a result, this can lead to low levels of learning
engagement (Bawa, 2016; Dumford and Miller, 2018) and motivation
(Ferrer etal., 2022; Fini, 2009; Mamolo, 2022; Wang and Lehman,
2021) among learners.
e need to tailor the learning experience to the individual learner
has been observed, mentioned, and studied many times in the current
literature (Klašnja-Milićević etal., 2011; Mutlu-Bayraktar etal., 2019;
Tekin etal., 2015; Wu etal., 2020). In educational psychology, the
concept of the Zone of Proximal Development (ZPD) developed by
Lev Vygotsky draws the theoretical foundations that support
personalized learning (Chaiklin, 2003; Tetzla etal., 2021; Vygotsky
and Cole, 1978). is concept emphasizes the need to understand that
each learner is at a dierent point in their cognitive development.
According to Vygotsky, the ZPD represents the set of tasks or skills
that a learner cannot yet perform alone but can perform with
assistance (Vygotsky and Cole, 1978). Learning is not encouraged by
tasks that are too simple or already within the scope of our current
abilities, leading to a state of boredom (Vygotsky and Cole, 1978).
Conversely, no learning occurs when tasks are overly complex and
frustrating tasks that exceed our abilities (Vygotsky and Cole, 1978).
us, maintaining a learner’s ZPD provides the ideal level of challenge
to promote growth and development, which can befurther enhanced
by personalized support and guidance to improve academic
performance over traditional “one-size-ts-all” teaching methods
(Alamri etal., 2021).
Complementary to the ZPD, the concept of cognitive load is
important for understanding and personalizing learning experiences
(Mutlu-Bayraktar etal., 2019; Sweller, 2020; van Merriënboer and
Ayres, 2005). Cognitive Load eory (CLT) examines human
cognitive architecture and provides insight into how learners process
and retain information in memory (Curum and Khedo, 2021; Sweller,
1988; Sweller etal., 1998; Wouters etal., 2008). is theory considers
the interplay between the working memory’s limited capacity and
long-term memory (Kalyuga and Liu, 2015; Mutlu-Bayraktar etal.,
2019). It denes cognitive load as the mental workload required to
perform a learning task and emphasizes the importance of managing
the mental eort required for eective learning (Kalyuga and Liu,
2015; Zhou etal., 2017a). us, performing a learning task requiring
too much or too little mental eort will lead to less-than-optimal
learning experiences and poor performances (De Jong, 2010). In a
CBL environment, ZPD can serve as a tool to tailor educational tasks
and support to suit the learner’s abilities, helping maintain cognitive
load at an optimal level while learning. Unfortunately, current CBL
environments only consider the learner’s perceived cognitive load as
a global design consideration, disregarding their objective cognitive
state evolution to fully tailor instructions to their abilities (Gerjets
etal., 2014; Sweller, 2020). One solution to this problem is the real-
time measurement of cognitive load through the electrical activity of
the brain using an Electroencephalogram (EEG)-based Brain-
Computer Interface (BCI) system.
BCIs facilitate direct communication between the brain and
computers by converting the brain’s electrical signals into computer
commands (Gao etal., 2021; Lotte etal., 2018; Zander and Kothe,
2011). Initially created to assist individuals with disabilities in
controlling external devices (Värbu etal., 2022), BCIs now extend to
passive systems that monitor cognitive states, such as attention,
fatigue, engagement, and cognitive load (Zander and Kothe, 2011),
enhancing cognitive functions through self-regulation and
neurofeedback (Birbaumer et al., 2009). ese systems provide
feedback based on brain activity changes, forming a closed
biocybernetic loop (Krol and Zander, 2017). BCIs potentially oer
tailored learning experiences in education by adjusting educational
content based on real-time brain activity analysis.
us, the purpose of this study is to investigate whether the use of
a neuroadaptive interface would provide an optimal learning
experience and increase learning gains with the following research
question: “Does adapting the pace of information presentation to the
learner’s real-time cognitive load using an EEG-based passive BCI
enhance the learning experience?.” Specically, we developed an
EEG-based BCI system that adapts the speed of information
presentation on the Interactive User Interface (IUI) according to the
real-time cognitive load of the learners. Wecreated a memory-based
learning task following the ZPD theory to test our BCI system. e
dynamic adaptive measures of our BCI are designed to help learners
manage their cognitive load and stay within their ZPD for an optimal
learning experience. Wedene an optimal learning experience as the
intersection of increased learning gains, self-perceived cognitive
absorption and satisfaction, and reduced self-perceived
cognitive workload.
Furthermore, the limited research on the use of BCI in education
fails to account for the impact of motivation on adaptation. While it
is established that motivation inuences the cognitive eort invested
in a learning task (Paas etal., 2005), there is a dearth of information
on this topic in the context of BCI-based learning. Wealso aim to
investigate if the addition of a motivational factor while using the BCI
would enhance the learning experience with the following research
question: “To what extent is motivation a necessary condition for
eective BCI adaptation?”
To the best of our knowledge, our study is the rst of its kind,
combining a novel BCI system and a memorization-based learning
task developed following the ZPD theory. Our research stands out
as very few papers study neuroadaptive interfaces in a CBL context.
Existing papers on the topic have used BCIs to monitor dierent
cognitive states (Andreessen etal., 2021; Marchesi and Riccò, 2013;
Zammouri etal., 2018; Zhou etal., 2017b), detect and react to error
potentials (Butteld et al., 2006; Spüler et al., 2012), to adjust
dierent interface parameters, such as task diculty or content type
(Eldenfria and Al-Samarraie, 2019) or provide user cognitive state

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 03 frontiersin.org
feedback (Verkijika and De Wet, 2015). In contrast, weemploy a
BCI system that uses real-time data to estimate and classify
cognitive load to adapt the speed of information presentation on
the interface.
e remainder of this manuscript is organized as follows. Werst
present related literature and the development of the hypotheses.
We then present the materials and methods used in this study,
including core aspects of developing our BCI system. Wethen present
our data analysis and study results. Findings are interpreted within the
discussion section. Finally, the article concludes with a short
conclusion encompassing limitations and future research avenues.
2 Related work
2.1 The need for individual learning paces
within the zone of proximal development
e ZPD theory suggests that all students have dierent learning
needs and abilities, therefore dierent ZPDs. us, within ZPD, each
student assimilates and processes new information or acquires abilities
dierently; some learners need more time and eort than others to
learn successfully (Hedegaard, 2012).
Studies have shown that in order to increase information retention
and promote optimal learning experiences, learning pace must
beadjusted and personalized to each student (Najjar, 1996; O'Byrne
and Pytash, 2015; Shemshack and Spector, 2020). For example, Hasler
etal. (2007) investigated the dierences between imposed system-
paced and personalized learner-paced groups on primary school
students. ey found that self-perceived cognitive load was lower and
test performance was higher when students used the learner-paced
system, which suggests that allowing students to control their own
learning pace may improve learning outcomes. Andreessen etal.
(2021) also investigated the eect of text diculty and text
presentation speed in a reading task on self-perceived mental
workload. Some texts, varying in diculty, were presented at the
reader’s pace, and some were presented at a 40% faster pace. Cognitive
load predicted values and subjective mental workload experienced
were signicantly higher when learners read at a fast-imposed speed.
In short, these studies demonstrate the importance of adapting
learning tasks, educational content, and instructional strategies to
each learner’s learning pace to promote an optimal learning
experience. ese studies also suggest that CBL environments
facilitate the personalization of learning methods and processes.
2.2 Personalizing computer-based learning
environments
CBL has created new opportunities for personalized learning in
the digital era. Personalizing learning through CBL can help address
each learner’s diverse learning needs by adapting instructional
materials to their learning pace and progress, which can help optimize
the ratio of challenge to support explained by the ZPD to suit
each learner.
Recent CBL environment studies rely on users’ personal and
learning data to create algorithms that personalize the learning
experience. For example, Xiao etal. (2018) developed a personalized
system that recommends learning materials based on an algorithm
combining the student’s learning path and interests. Results from the
pilot testing indicated that their system increased the learners’ learning
outcomes and satisfaction levels. El-Sabagh (2021) developed an
online learning environment that suggests content based on the
student’s learning style and adapts the modules based on behavioral
data (learning activities, errors, navigation). ey found that the
participants who used the adaptive learning environment had better
overall performance scores and higher reported engagement levels
than those who did not. Ku and Sullivan (2002) also developed an
adaptive learning system that adapts mathematical questions based on
the learner’s interests (favorite foods, sports, etc.) and discovered that
the system enhanced the students’ learning achievement and positively
aected their learning attitude. Finally, Tekin etal. (2015) developed
eTutor, a personalized online learning platform, that learns the best
order in which to deliver instructional materials with an algorithm
based on the learner’s preferences and needs, and uses their feedback
input on previously presented instructional contents (such as exam
scores and time spent on a course) to adapt the educational material.
ey found that their system improved performance on assessments
and achieved signicant savings in the amount of time that students
spent learning.
ese studies have demonstrated that adaptive CBL environments
can positively impact the learner’s learning experience. However, their
assessment methods do not account for the learner’s real-time
cognitive load, which can substantially aect learning eectiveness
and eciency (Sweller, 1988, 2020).
2.3 Cognitive load and measurement
approaches
e CLT postulates the importance of minimizing the mental
eort associated with the processing of the instructional design or the
learning environment (Curum and Khedo, 2021; DeLeeuw and Mayer,
2008) that is unrelated to the learning itself (Extraneous Load) and
managing the level of complexity of both the learning material and the
learning task itself (also known as Intrinsic Load) (Sweller, 2010), in
order to reduce the overall cognitive load and thereby optimize the use
of working memory resources [known as Germane Load (Debue and
van de Leemput, 2014) or Germane Processing (Sweller etal., 2019)].
Werefer to this sweet spot as the “Goldilocks Zone” (Karran etal.,
2019), where the overall cognitive load is optimized to enhance the
learning process and increase performance.
ZPD and cognitive load are closely linked concerning the
personalization and optimization of learning experiences. Learning
tasks that align with a student’s ZPD are less likely to overwhelm them,
helping to reduce their Extraneous Load (Schnotz and Kürschner,
2007). In addition, instruction tailored to a learner’s ZPD facilitates
the learning and minimizes their Intrinsic Load (Schnotz and
Kürschner, 2007). us, the ZPD makes it possible to evaluate the
learner’s cognitive abilities to avoid cognitive overload and underload,
leading to poor learning outcomes (Paas etal., 2004).
It is essential to measure and assess the cognitive load of learners
to adjust their learning environments and enhance their learning
experiences and outcomes. Today, self-reported measures remain the
most used method to measure cognitive load in the research and
development of various educational technology tools as they oer the

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 04 frontiersin.org
learners’ perspectives on their experience (Anmarkrud etal., 2019;
Brunken etal., 2003; Mutlu-Bayraktar etal., 2019). However, they
cannot objectively and precisely capture and quantify the amount of
mental work expended during the learning process (Mutlu-Bayraktar
et al., 2019). Self-perceived measures also rely on the learners’
subjective awareness and perceptions, which involve a deeper
reection and thought process about their learning experience (Ayres,
2006). Learners must reect upon their learning experience,
considering the cognitive eort and mental processes involved,
inuenced by their level of metacognitive awareness. While subjective
measures oer insights into the perception of cognitive load, they do
not fully capture the learner’s evolving cognitive state, which is
necessary to tailor instructions to their abilities. Utilizing physiological
measurement tools such as eye movement data, hormone levels, heart
rate variability, and brain activity (Riedl and Léger, 2016) can provide
a more precise, reliable, valid and complementary continuous
cognitive load assessment (Brunken etal., 2003).
Among the various tools available for brain imaging, EEG is one
of the most used due to its non-invasive, cost-eective, convenient,
accessible features and high temporal resolution (Abiri etal., 2019;
Antonenko etal., 2010). EEG measures voltage uctuations in cortical
activity, which can beused to assess and infer mental states. Dierent
cognitive processes are associated with variations in brainwave
patterns, specically frequency, amplitude, synchronization between
neural networks, and Event-Related Potentials (ERPs) in response to
stimuli (Riedl and Léger, 2016). Previous research on cognitive load
suggests that theta (θ, 4–7 Hz) and alpha (α, 8–12 Hz) oscillations are
associated with task diculty, with alpha activity becoming
desynchronized (or decreased) and theta activity becoming
synchronized (or increased) as task diculty increases (Antonenko
etal., 2010; Gevins and Smith, 2003; Klimesch, 1999; Stipacek etal.,
2003). Dynamic changes in alpha activity would mainly occur in the
brain’s posterior regions, while changes in theta activity would mainly
occur in the brain’s frontal regions (Cavanagh and Frank, 2014;
Tuladhar etal., 2007). Prior research used a visuospatial working
memory task to explore whether variations in brain activity
synchronization within and between the frontal and parietal regions
stem from diering central executive demands (Klimesch etal., 2005).
e ndings indicated that activity synchronization between these
areas’ mirrors working memory’s executive functions: increased
executive load leads to reduced anterior coupling in the upper alpha
range (10–12 Hz) and heightened theta synchronization between
frontal and parietal regions.
2.4 Brain-computer interfaces
BCIs enable direct brain-to-machine communication and
interaction, allowing users to manipulate and engage with technology
(Gao etal., 2021; Lotte etal., 2018; Zander and Kothe, 2011). BCI
research has gained much popularity in recent years due to its
potential medical applications (Gu et al., 2021), such as for
neurorehabilitation in brain injury, motor disability and
neurodegenerative diseases (Abiri etal., 2019; Chaudhary etal., 2016;
Daly and Wolpaw, 2008; Pels etal., 2019; Vansteensel etal., 2023),
detection and control of seizures (Liang etal., 2010; Maksimenko
etal., 2017), and improvement of sleep quality and automatic sleep
stages detection (Papalambros etal., 2017; Phan etal., 2019). Several
studies have also looked at non-clinical applications, such as video
games (Ahn etal., 2014; Kerous etal., 2018; Laar etal., 2013; Labonte-
Lemoyne etal., 2018; Lalor etal., 2005; Lécuyer etal., 2008), marketing
and advertisement (Bonaci etal., 2015; Mashrur etal., 2022; Tadson
etal., 2023), neuroergonomics and smart environments (Carabalona
etal., 2012; Kosmyna etal., 2016; Lin etal., 2014; Tang etal., 2018),
and work monitoring and safety (Aricò etal., 2016; Demazure etal.,
2019; Demazure etal., 2021; Karran etal., 2019; Roy etal., 2013;
Venthur etal., 2010). A BCI is classied as a neuroadaptive interface
(Riedl etal., 2014) when real-time adaptations occur on an interface
presented on a computer.
Most BCIs use EEG to acquire brain signals (Lotte etal., 2018).
Depending on the type of research conducted, EEG-based BCIs can
beinvasive (with electrodes placed directly on the surface of the brain)
or non-invasive (with electrodes placed on the scalp of the subject)
(Abiri etal., 2019). Invasive EEG-based BCIs have the advantage of
directly measuring higher-quality brain signals, reducing external
interference (Daly and Wolpaw, 2008). However, they require surgery
to insert and remove the electrodes, exposing patients to several
potential complications (Daly and Wolpaw, 2008; Värbu etal., 2022).
In contrast, non-invasive EEG-based BCIs measure brain activity
using electrodes placed on the scalp. e major drawback is that these
electrodes are subject to several factors that aect the quality of the
recorded signal, such as external noise, a weaker electrical signal, and
even the physical movements of the subject (Padeld etal., 2019).
Nevertheless, non-invasive EEG-based BCIs remain more popular due
to their noninvasiveness while providing high temporal resolution and
a low cost (Abiri etal., 2019; Cohen, 2017; Dimoka etal., 2012; Lotte
etal., 2018; Värbu etal., 2022).
In general, brain signals are typically rst acquired with an EEG
(Lotte etal., 2018), which are then processed through a series of steps,
including data preprocessing, feature extraction and signal
classication (Padeld etal., 2019), before nally being interpreted by
the BCI and used for its purpose (Abiri etal., 2019; Lotte etal., 2018).
ere are three main BCI paradigms: active, reactive, and passive
(Table1). Active paradigms allow users to directly control the system
by deliberately controlling their brain activity (Ahn etal., 2014;
Angrisani etal., 2021; Zander and Kothe, 2011; Zander etal., 2009).
For instance, users can employ mental imagery to imagine motor
movements, allowing the system to replicate the intended action on
a screen or with an external device, such as a mechanical arm
(Steinert etal., 2019). In reactive paradigms, specic brain activity
initiates predetermined actions from the system in response to
external stimuli (Ahn etal., 2014; Wang etal., 2019; Zander and
Kothe, 2011). Brain reactivity measured following external stimuli
is associated with a specic command from the system, making this
type of BCI very specic and ecient (Dehais etal., 2022). For
example, Chen et al. (2017) used Steady-State Visual-Evoked
Potentials (SSVEP) to develop a reactive BCI in a visual navigation
task. SSVEPs were detected by the BCI when participants were
looking at the sides of a ickering square in the middle of the screen,
which allowed them to control the direction of the cursor. Finally, in
a passive paradigm, brain activity is continuously monitored to
dierentiate or quantify mental states without user control,
providing feedback as a system response. For example, Karran etal.
(2019) developed an EEG-based passive BCI to measure and
monitor users’ sustained attention in a long-duration business task.
e system’s feedback consisted of countermeasures in the form of

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 05 frontiersin.org
color gradients representing the participant’s sustained attention
level and alerts when sustained attention was low as forms of system
feedback to maintain sustained attention at an optimal level and
improve performance.
Passive BCIs have garnered signicant attention recently,
especially for implementing closed-loop adaptations (Krol and
Zander, 2017). In a passive closed-loop BCI, real-time brain activity
and adaptive system actions continuously inuence each other as part
of a biocybernetics loop (Ahn etal., 2014; Krol and Zander, 2017;
Pope etal., 1995; Roy etal., 2013; Zander and Kothe, 2011). is
dynamic cycle begins when an assessed brain state triggers an adaptive
response from the system. e system then provides feedback or
adjusts the content to alter the current brain state, and so forth (Krol
and Zander, 2017). e aforementioned study by Karran etal. (2019)
is an example of a closed-loop BCI, as the system continuously
monitors sustained attention and provides feedback according to the
level measured to inuence the user to increase their sustained
attention. is biocybernetics loop continued until the end of
the experiment.
2.5 Brain-computer interfaces in
educational contexts
e application of BCIs in diverse settings demonstrates their
innovative potential to enhance learning outcomes and empower
learners through novel interactions with educational content.
However, research on using BCIs in educational contexts is limited
and inconsistent (Xia etal., 2023). Previous studies have primarily
employed passive BCIs to achieve mental state assessments of users
as they learn and interact with educational interfaces, subsequently
personalizing learning according to the data collected (Krol and
Zander, 2017). For example, Apicella et al. (2022) developed a
wearable EEG-based system to detect and classify students’
cognitive and emotional engagement during learning tasks,
leveraging brain signals to optimize adaptive learning platforms in
real-time. Engagement was measured using EEG signal analysis
through a Filter Bank and Common Spatial Pattern (CSP) method,
followed by classication with a Support Vector Machine (SVM).
e task involved a Continuous Performance Test (CPT) to
modulate cognitive engagement, while emotional engagement was
inuenced by background music and social feedback. e system
achieved classication accuracies of 76.9% for cognitive and 76.7%
for emotional engagement. In addition, previous research on
cognitive load and adaptive educational interfaces has mainly
focused on the complexity of the educational material and the
instructional guidance presented to the learner (Kalyuga and Liu,
2015; Mutlu-Bayraktar etal., 2019; Petko etal., 2020). ese gaps in
the literature have recently prompted researchers to investigate the
transformative potential of passive closed-loop BCIs in
learning contexts.
For instance, Yuksel etal. (2016) created a passive closed-loop BCI
called Brain Automated Chorales (BACh), which adjusts the diculty
level of piano learning material according to cognitive workload
measurements obtained through functional near-infrared
spectroscopy (fNIRS). Adaptive measures of the system depended on
learners’ cognitive workload throughout both the training and
learning tasks, which were classied using a machine learning
algorithm. e results suggest that the learners’ playing speed and
performance accuracy improved when learning piano with the BACh
system. Additionally, the learners reported a better learning experience
with the system and noted that diculty levels were appropriately
adjusted. Additionally, Walter etal. (2017) designed a closed-loop
EEG-based BCI that measures cognitive workload in real-time to
adapt the diculty of arithmetic problems presented in an online
learning environment. Cognitive workload classications were
separated into three diculty levels based on workload state
predictions derived from a pre-trained regression model to determine
the optimal range of cognitive workload for learning. eir ndings
demonstrated that participants who completed the experiment with
the adaptive instructions achieved greater learning gains than those
who completed the experiment without adaptivity. However, this
dierence was not statistically signicant. Finally, Kosmyna and Maes
(2019) created AttentivU, an EEG-based passive closed-loop BCI that
measures engagement in real-time and triggers haptic feedback
(vibrations from a scarf worn by the learner) when a drop in
engagement is detected. e system used the engagement index
proposed by Pope etal. (1995), which calculated the average power of
theta, beta and alpha frequency components derived from Power
Spectral Density to return a smoothed engagement index every 15 s.
e two studies conducted with AttentivU yielded results indicating
that haptic biofeedback driven by BCI redirected learners’ engagement
to the task, resulting in enhanced performance on comprehension
tests. ese studies demonstrate the feasibility of closed-loop BCI
systems within educational contexts to adapt and personalize learning
to each learner.
TABLE1 Overview of brain-computer interface (BCI) paradigms: control types, user involvement, applications and advantages.
Control type User involvement Common applications Advantages
Active BCI User-driven, conscious
control of brain activity
High: deliberate modulation of
brain signals by the user (e.g.,
motor imagery).
Neuroprosthetics, motor control
(e.g., robotic arm).
Fine-tuned control for specic
tasks, useful for disabled users
needing direct control.
Reactive BCI Stimulus-driven, system
reacts to external stimuli.
Medium: passive response to
external stimuli (e.g., Steady-State
Visual Evoked Potentials (SSVEP),
P300).
Speller systems, attention-based
interfaces.
Ecient, system commands are
linked to specic brain responses
to stimuli.
Passive BCI System-driven, monitors
brain states without user
control.
Low: no direct user control;
continuous monitoring of
spontaneous brain activity.
Cognitive workload assessment,
fatigue monitoring, adaptive
systems.
Non-intrusive, ideal for
monitoring and real-time
adaptation to mental states.

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 06 frontiersin.org
e aim of the current study is to investigate the eects of an
EEG-based passive closed-loop BCI on the learning experience in a
memory-based learning task and contribute to the literature regarding
the eects of closed-loop passive BCI on learning outcomes.
2.6 Hypotheses development
Our study aims to answer the following research question: “Does
adapting the pace of information presentation to the learner’s real-time
cognitive load using an EEG-based passive BCI enhance the learning
experience?.” Wehypothesize that (H1) “neuro-adaptivity enhances the
learning experience compared to the absence of neuro-adaptivity”
(Figure1). is study denes the learning experience as a combination
of objective and subjective measures of cognitive load and emotional
state, specically focusing on learning gains, perceived mental
workload, perceived cognitive absorption, and satisfaction.
Learning gains in this context represent an objective measure of
the knowledge learned and memorized throughout the experimental
task, allowing an assessment of the impact of the BCI on learning.
Prior research suggests aligning learning speed with cognitive load
can enhance eciency and eectiveness (Petko et al., 2020).
Wepropose that neuro-adaptivity leads to greater learning gains by
optimizing the learning pace to the learner’s cognitive load. us,
wehypothesize (H1a) that neuro-adaptivity leads to more signicant
learning gains compared to the absence of neuro-adaptivity (Figure1).
Additionally, understanding how learners perceive and estimate
their mental workload while working with and without the BCI, is
necessary for evaluating the learning experience. Perceived mental
workload refers to the perceived mental eort required to complete
the learning task and its impact on the experience (Hancock and
Meshkati, 1988), where higher perceived mental workload translates
into a less optimal learning experience (Sweller, 1994). erefore,
wehypothesize that (H1b) “neuro-adaptivity reduces perceived mental
workload compared to the absence of neuro-adaptivity” (Figure1).
Derived from Csikszentmihalyi’s theory of ow
(Csikszentmihalyi, 1975; Csikszentmihalyi, 2014), cognitive
absorption is described as a state of total immersion when
performing a task, characterized by high levels of engagement and
focus (Agarwal and Karahanna, 2000). Previous studies have shown
that higher levels of cognitive absorption while completing CBL
tasks lead to higher satisfaction levels and better-perceived ease of
use and usefulness of the learning tool (Saadé and Bahli, 2005;
Salimon etal., 2021). erefore, wehypothesize that (H1c) “neuro-
adaptivity generates a higher self-perceived cognitive absorption level
than the absence of neuro-adaptivity” (Figure1).
Learner satisfaction reects the degree to which learners feel
engaged, satised, and fullled with their learning experiences
(Martin and Bolliger, 2022; Wickersham and McGee, 2008). Previous
research has shown that learner satisfaction leads to better learning
outcomes (Martin and Bolliger, 2022). erefore, we hypothesize
(H1d) that “neuro-adaptivity generates a higher level of perceived
FIGURE1
Conceptual framework illustrating the eects of neuro-adaptivity and motivation on learning outcomes.

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 07 frontiersin.org
satisfaction with the learning experience compared to the absence of
neuro-adaptivity” (Figure1).
Furthermore, weaim to examine the role of motivation, both
intrinsic and extrinsic, in the learning experience during BCI
utilization. Numerous studies have demonstrated the importance of
motivation in achieving academic success, notably in CBL
environments (Hu etal., 2016; Lepper and Malone, 2021; Nikou and
Economides, 2016). To aid in this examination, weask the following
research question: “To what extent is motivation a necessary condition
for eective BCI adaptation?”
In general, learners are more likely to beactively engaged and
motivated when the learning experiences provided are specic to their
ZPD (Shabani et al., 2010; Vygotsky and Cole, 1978). Self-
determination theory (SDT) investigates the motivations of
individuals in varying social contexts and situations. It identies two
types of motivation: intrinsic and extrinsic (Ryan and Deci, 2000a,b).
When learners are intrinsically motivated, they will learn naturally,
usually with interest and enjoyment, because of the benets that the
subject matter can bring (Ryan and Deci, 2000a,b). Whereas, extrinsic
motivation occurs when learners compel themselves to learn to obtain
a reward or avoid consequences (Ryan and Deci, 2000a,b). Extrinsic
incentives such as money or prizes have been demonstrated to
enhance learning performance (Schildberg-Hörisch and Wagner,
2020) by improving attention (Anderson, 2016; Small etal., 2005),
eort (Schwab and Somerville, 2022), and working memory (Wimmer
and Poldrack, 2022) and can motivate students to remain interested,
engaged, and dedicated to their learning, resulting in greater learning
outcomes (Festinger etal., 2009; Gong etal., 2021; Rousu etal., 2015).
ese ndings suggest that extrinsic motivators can support
intrinsic motivation. erefore, wewill utilize extrinsic motivation in
the form of a nancial incentive to help answer our research question.
We hypothesize that (H2) motivation moderates the eect of
neuroadaptation by increasing its eectiveness and perception of an
optimal learning environment when compared to the neuro-adaptive
interface alone (Figure1). More precisely, wehypothesize that (H2a)
adding motivation to neuro-adaptivity helps to achieve greater learning
gains compared to neuro-adaptivity alone; (H2b) adding motivation to
neuro-adaptivity reduces perceived mental workload compared to
neuro-adaptivity alone; (H2c) adding motivation to neuro-adaptivity
generates a higher level of perceived cognitive absorption than neuro-
adaptivity alone; (H2d) adding motivation to neuro-adaptivity
generates a higher level of self-perceived satisfaction of the learning
experience compared to neuro-adaptivity alone (Figure1).
3 Materials and methods
3.1 Participants
Fiy-ve participants participated in our study (27 ± 7.92 years
old, 28 female), 36 university students, 19 took online classes or
training regularly for professional or personal reasons. All
participants were recruited by e-mail from our institution’s panel
database. Participants were included based on age greater than
18 years old, normal or corrected-to-normal vision, having no
history of neurological conditions, right-handedness, uency in
the French language, and high computer prociency. Handedness
was validated before the experiment with the Edinburgh
Handedness Inventory (Caplan and Mendoza, 2011), and all other
inclusion criteria were validated through the screening
questionnaire. Participants signed a consent form before
completing the study and were informed they could leave it
anytime. Participants were compensated 100$ (CAD) for their
participation. Our institution’s ethics committee approved the
study under certicate 2023–5,071.
3.2 Experimental design
3.2.1 Experimental conditions
We utilized a 3 × 2 (type of adaptation x motivation) between-
subject design. Participants were randomly assigned to a group prior
to data collection and kept unaware of experimental factors. In the
current study, conditions refer to type of Interactive User Interface
(IUI): Control (C) no adaptivity (n = 17), stimuli are presented at
predened intervals; Adaptive (A) without motivation (n = 22),
stimuli are presented at variable speeds based on a classication of
user cognitive load; Adaptive (AM) with motivation (n = 16), stimuli
are presented at variable speeds based on a classication of user
cognitive load in the presence of nancial motivation. For the
AM group to provide extrinsic motivation, participants were
informed that better overall task performance resulted in more
entries in a $200 Visa prepaid gi card prize draw. To conform with
ethical principles, all participants, regardless of experimental
condition, received the same number of entries for the prize draw
when the study concluded.
3.2.2 Phase one: calibration
As illustrated in Figure2, the rst part of the calibration phase
consisted of a 90s baseline task used for post-hoc analyses, where
participants had to stare at a black square in the middle of a grey
screen. e second part of the calibration phase consisted of an n-back
task to estimate personal threshold values of high and low cognitive
load. ese thresholds were then integrated into the BCI model to
personalize the classier’s thresholds and limits (Sections 3.2.4 and
3.3.2). is task was performed regardless of condition.
e n-back task was selected due to its popularity for manipulating
memory load, which can serve as a proxy for cognitive load (Brouwer
etal., 2012; Grimes etal., 2008; Wang etal., 2016) and its similarity to
the learning task, which requires memory and recall of visual stimuli.
In the n-back task, participants must assess whether each stimulus in
a sequence corresponds to the stimulus presented n items earlier
(Hogervorst etal., 2014). As n increases, the n-back task becomes
more challenging, requiring more cognitive resources. A four-minute
n-back task was administered in two parts: a 2-min 0-back task to
assess low cognitive load and a 2-min 2-back task to assess high
cognitive load, separated by a short break of 30 s. Each stimulus
(letter) was presented for one second, followed by a two-second
intertrial interval for both tasks, resulting in a new letter being
presented every three seconds, totaling 40 iterations.
3.2.3 Phase two: learning task
One of the most frequent learning tasks in higher education
involves memorizing course material for exams and practical
applications due to the sheer quantity of information that must
be learned within a limited time frame. To test our hypotheses,

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 08 frontiersin.org
we adapted an existing constellation memorization learning task
(Riopel etal., 2017).
Star constellations were chosen as the learning topic for two
reasons. First, university students typically possess low prior
knowledge about the subject. Second, even the most knowledgeable
individuals easily encounter unfamiliar material. e task required
participants to select the correct name of a constellation from three
options associated with an image of one of the 88 constellations. e
purpose was to examine the learning, forgetting, and spacing curves
in online learning. is allowed us to design a valid task that could
promote learning while inducing changes in cognitive load.
As indicated in Figure3, participants were instructed to memorize
as many constellations as possible by associating the presented
constellation image with its corresponding name from a choice of four
multiple-choice answers. e correct answer (feedback) was displayed
aer each question, regardless of whether it was answered correctly or
not. Previous research has indicated that providing the correct answer
to a question, irrespective of whether it was answered correctly, is
essential in enhancing the retention of information and avoiding
future mistakes (Butler etal., 2008; Kulhavy, 1977). e instructions
remained the same throughout the learning task, which contained
four blocks (i.e., trials) of questions, separated by short breaks of 30 s
(see Section 3.2.5). Participants were required to memorize 32
constellations, each presented twice per block. e sequence of
constellation presentation was pre-randomized before data collection
and remained the same for all participants. However, the correct
answer’s position among the four multiple-choice options and the
three incorrect constellation names were randomized.
3.2.4 Model of adaptivity and cognitive load
classifications
e model of adaptivity used in our study was adapted from
Karran etal. (2019) who conceived of an adaptive model of sustained
attention, in which two thresholds denote the chance of failure, an
upper so limit beyond which chances of failure increase, and a lower
hard limit beyond which failure is certain, the model is such that
adaptive countermeasures are provided to keep a user of the BCI
within the upper and lower bounds in what they term the “goldilocks”
zone, i.e., neither to high nor too low. Wechose this model because it
was easily adapted to replace sustained attention with cognitive load,
while keeping all thresholds the same. In the current study, weinverted
the limits such that the upper limit represents cognitive overload and
FIGURE2
Schematic representation of the n-back task used in the calibration task (phase 1).
FIGURE3
Example of a constellation from the learning experiment, presented on the interface.

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 09 frontiersin.org
the certainty of failure, and the lower limit represents to little cognitive
load and an increased chance of failure through inattention or
boredom. e “goldilocks” zone represents the ZPD, which promotes
an optimal cognitive load level, which is not too high or too low,
through uid and dynamic adaptations to enhance learning gains
over time.
EEG analysis between the 0-back and the 2-back tasks controlling
the False Discovery Rate (FQR, q = 0.05) of 17 pre-tests demonstrated
a signicant decrease in alpha-band activity within the parietal,
occipital, and right temporal regions. However, the same analyses with
a Bonferroni correction suggested a signicant reduction in alpha-
band (α) activity at the P7 electrode. Consequently, weexclusively
used the P7 electrode when computing the cognitive load index,
which aligns with the current literature (see Section 2.3). Weused an
index based on average alpha-band power in the parietal cortex
(electrode P7) during 6-s sliding windows with no overlap to calculate
the cognitive load.
,α
=
current i
CL P
Where
current
CL
represents a new real-time index value, i.e., the
current cognitive load level, by calculating the alpha-band power
activity during the ith 6-s sliding window, denoted by
á,i
P
.
As described in Section 3.2.2, the n-back task was used to
determine baseline cognitive load thresholds. Specically, cognitive
load averages for the 0-back and 2-back tasks were calculated
separately using the cognitive load index. is resulted in the creation
of two thresholds, which represent “low average” and “high average”
cognitive load. In addition, the average cognitive load for the entire
n-back task was calculated.
1
02
and CL =
=∑
N
current
i
back back
CL
CL
N
02
n
2
+
=back back
back
CL CL
CL
Where
0back
CL
and
2back
CL
denote the average cognitive load for
the 0-back or the 2-back task, respectively. N represents the total
number of 6-s sliding windows during the task, used to calculate the
average of the task.
current
CL
represents the real-time cognitive load
level, i.e., a new real-time index value, calculated with the cognitive
load index. Finally, the average cognitive load level is calculated using
the 0-back and 2-back task thresholds, denoted by
nback
CL
.
e real-time index values were stabilized during the learning
experiment using a 60-s sliding window that dynamically adjusted the
average cognitive load over time. In other words, decisions on
cognitive load classications were made every 6 s based on the index
compared with a moving average of the previous 60 s or the last 10
data points. is ensured that the classication would adjust to
changes in the user’s cognitive state throughout the experiment.
Additionally, analysis of the 17 pre-tests indicated a 125% increase in
the amplitude of the alpha-band signal during the learning task
compared with the n-back task. ese results suggest that the
thresholds should be 1.25 times higher than the average values
obtained in n-back. erefore, the resulting cognitive load value
exceeding the “high average” threshold would result in a classication
as “2” in the BCI system, indicating a high cognitive load level.
Conversely, a resulting cognitive load value below the “low average”
threshold would classify as “0” in the BCI system, indicating a low
cognitive load level. Finally, when the resulting cognitive load value
fell between the “high average” and “low average” thresholds, it would
be converted to a “1” classier in the BCI system, indicating an
optimal level of cognitive load.
i9
i
10
= −
=∑i
current
jCL
MA
0
n
0 1.25

=××



back
i
back
CL
Class MA CL
2
n
2 1.25

=××



back
i
back
CL
Class MA CL
Where
current
CL
represents the real-time cognitive load value,
calculated with the index. erefore,
i
MA
represents the moving
average of the last ten cognitive load values at time i. e factor of 1.25
represents the threshold adjustment according to the results obtained
in the pre-tests.
3.2.5 Adaptive rules of the interface and
specifications
During the learning task, the adaptive Intelligent User Interface
(IUI) modulated the information delivery speed (see Figure 4).
Specically, upon receiving high cognitive load classications (“2”),
the interface slowed information delivery, aording participants
extended time for question response and correct answer processing.
Conversely, low cognitive load classications (“0”) triggered an
increase in delivery speed, reducing the response and correct answer
display time. No adjustment was made for classications of average
cognitive load (“1”), indicating optimal cognitive load. Following ZPD
theory, weposited that these time adaptations would allow the learner
to remain in their ZPD, leading to better learning outcomes. us, the
baseline time window for displaying constellation questions and
feedback was 5 s. Based on pre-test results, adjustments were made in
1-s increments within a 3 to 8-s range per item. Pre-tests revealed that
presentations over 8 s diminished response eciency and signicantly
lowered engagement, focus, and interest, aligning with existing
research ndings (Beck, 2005; Chipchase etal., 2017). e minimum
time was set at 3 s to prevent the BCI system from getting confused
between the brain’s processing of new information and high cognitive
load levels (Anderson etal., 2011; Rosso etal., 2001; Vijayalakshmi
etal., 2015). Finally, the constellation question and the feedback were
presented for the same duration to ensure adequate time for
participants to respond and process the correct answer.
Figure5 illustrates the learning task, which was structured into
four blocks, interspersed with 30-s intervals. In the C group, question
and feedback pacing remained constant across all blocks, adhering to

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 10 frontiersin.org
a 5-s baseline. For the A and AMgroups using the adaptive IUI,
information delivery rates in the second and third blocks were
modulated based on cognitive load classications from the BCI; no
adaptation was applied in the rst and last blocks to assess the eect.
To facilitate participant re-engagement post-breaks, the initial 30 s (or
rst three constellations) of the adaptive blocks maintained the
baseline delivery speed of 5 s for both questions and feedback.
No adaptation occurred while a constellation and its correct
answer were displayed. is way, if a high or low cognitive load
classication were received during this period, any change in the
speed of information provision would only aect the next constellation
to avoid confusing the learner. To prevent unnecessary stress during
short response times and loss of interest or focus during longer
response times, a countdown timer was clearly displayed below the
multiple choices to assist participants in managing their expectations
(Ghafurian etal., 2020). Finally, neither correct nor incorrect answers
inuenced the speed of information presentation, only the cognitive
load classication.
3.3 Apparatus
3.3.1 Interactive user interface
e constellation learning IUI was presented to the participants
on a 22-inch LED monitor with a resolution of 1,680 x 1050p and a
refresh rate of 60 Hz, running on a Windows PC and equipped with a
keyboard and a mouse. Participants were seated approximately 25
inches from the computer screen. e IUI was developed as a dynamic
Web application with AngularJS and was presented on Google
Chrome in full-screen mode. A rule engine was implemented in the
Web application to enable switching between the experimental
(adaptive IUI) and control (regular IUI) conditions. Adaptive rules
(see Section 3.2.5) were stored in a JSON le and loaded automatically
upon selection of the experimental condition. When either condition
was selected, a unique link was created for each participant that led to
the appropriate interface version, and placeholder database entries
were created to store the data. is data could beextracted directly
from the IUI as a JSON le for subsequent analysis.
3.3.2 The passive BCI
e BCI model was created using Simulink and MATLAB
(version R2021b, Mathworks, MA) with the g.HIsys environment
(g.tec medical engineering GmbH, Austria), which enables high-speed
online data processing. e BCI system ran on a Windows PC
operated by the researchers. Upon opening the BCI model, a folder
was created for each participant number to store EEG data. e
n-back task was integrated and directly accessible from the Simulink
model. Cognitive load thresholds derived from the n-back are stored
in the participant’s folder aer task completion for integration into the
BCI model.
e BCI system operated as a closed-loop mode, continuously
measuring cognitive load and adapting the speed of information
FIGURE4
Adaptive rules of the BCI system implemented in the experiment.
Block 1Block 2Block 3Block 4
30s30s 30s
C
(n=17)
A
(n=22)
AM
(n=16)
Adaptive learning
speed
Imposedlearning
pace
FIGURE5
The learning task: adaptivity of each block for each group.

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 11 frontiersin.org
presentation on the IUI (Figure6). e BCI acquires and processes
EEG signal, extracting features of alpha and theta band activity
from the P7 electrode. e extracted features are then used to
compute the cognitive load index. ese stabilized values are
compared to dynamic thresholds, resulting in a classication of
three potential levels: 0, 1, and 2. e classications are then
transmitted every 6 s using Lab Streaming Layer (LSL) to a Python
script. e script transmits the classications to the interface every
six seconds through WebSocket communication. Subsequently, the
classications are utilized in a rule engine to trigger the appropriate
adaptive actions.
While only activity within the alpha and theta bands of the P7
electrode was considered in the analysis and classication of cognitive
load, the BCI system monitored and stored brain activity from all 32
electrodes. e BCI stores the ltered P7 signals and the raw EEG data
separately, which can beretrieved for post-hoc analyses. Finally, the
BCI enables real-time visualization of EEG signals during the
calibration and learning tasks to monitor signal quality and
potential artifacts.
3.3.3 EEG real-time processing
Brain activity was continuously sampled using an active,
32-channel wireless and gel-based g.Nautilus Research EEG headset
(g.tec medical engineering GmbH, Austria) with g.Scarabeo electrodes
(Standard 10–20 System placement, see Figure7). e EEG amplier
was secured in a holder shell at the base of the cap and xed with
Velcro. e real-time sampling rate was set to 250 Hz and ltered
using bandpass (0.5 Hz – 30 Hz) and notch (58 Hz – 62 Hz) lters
applied in real-time. Each electrode was equipped with an amplier
to enhance signal quality, minimize artifacts, and reduce signal
degradation. e reference electrode was placed on the participant’s
right earlobe to aid in common-mode rejection.
3.4 Psychometric instruments
Questionnaires were administered to the participants using
Qualtrics (Qualtrics, Provo, UT) via anonymous links. Prior to
completion, participants were required to enter their participant
number, for later anonymous identication and analysis. Table 2
presents a summary of the questionnaires used in this study, including
the degree of internal consistency for questionnaires with multiple
items, which was assessed via Cronbach’s alpha (α).
3.4.1 Pre-test questionnaire
e pre-test questionnaire collected demographic information
and assessed participants’ prior knowledge and interest level in the
learning topic. First, the questionnaire requested participants to enter
their age (in years) and indicate the gender with which they identify
to. en, a simple Yes or No question evaluated the learner status
(whether the participant is a student). Answers were converted into
binary data, where 0 represented No, and 1 represented Yes. e prior
level of interest in the learning topic was assessed with a 10-point
Likert scale ranging from 1 “No interest” to 10 “Very interested.”
Finally, knowledge of learning topic was assessed using a 10-item
questionnaire adapted to the learning topic using a Likert scale
ranging from 1 “Strongly disagree” to 7 “Strongly agree” (Flynn and
Goldsmith, 1999). All items were averaged to create individual overall
scores, where the higher the scores, the higher the prior level of
knowledge. e internal consistency analysis revealed a questionable
Cronbach’s alpha value (α = 0.709). erefore, the second item was
discarded to increase the internal consistency to a higher, more
acceptable level (α = 0.772).
3.4.2 Post-test questionnaire
Perceived mental workload was evaluated aer the experiment
with the raw NASA-TLX questionnaire (Hart and Staveland, 1988),
BCI calibration (n-back)
Signal acquisition (EEG)Signal processing
CL index
CL thresholds
Closed-loopBCI
Signal acquisition (EEG)Signal processing CL index
Python scriptWebSocket
Web app (IUI)
LSL
Adaptive actions
Classification
Feedback
FIGURE6
Visual representation of the BCI system operation, with the calibration task (n-back).
FIGURE7
Electrode positioning of the EEG cap, P7 Indicated in blue and
reference electrode denoted as REF.

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 12 frontiersin.org
composed of six dimensions: mental demand, physical demand,
temporal demand, performance, eort, and frustration. A single
item represented each dimension. Participants were asked to
complete each item based on their learning experience. All
dimensions were measured using slidable cursors on a continuous
scale ranging from 0 to 100. Scores were rounded in post-hoc
analyses to t the questionnaire’s original calculations (Hart, 1986).
is allowed for an overall mental workload score to beobtained, as
well as individual observations of each dimension (Galy etal., 2018).
e initial Cronbach’s alpha (α = 0.689) calculation showed moderate
internal consistency. Weremoved the physical demand dimension
to achieve an acceptable alpha of (α = 0.715). Removing an item is
acceptable when using the raw NASA-TLX (Colligan etal., 2015;
Hart, 2006).
Perceived cognitive absorption was measured aer the experiment
using an adapted version of the Cognitive Absorption questionnaire
(Barki etal., 2008). is questionnaire covers the ve dimensions,
temporal dissociation, focused immersion, heightened enjoyment,
control, and curiosity, as described by Agarwal and Karahanna (2000),
to assess cognitive absorption, with three items per dimension. Items
were measured with 7-point Likert scales, ranging from 1 “Strongly
Disagree” to 7 “Strongly Agree.” An overall average score and an
average score of each dimension were calculated and interpreted with
high internal consistency Cronbach’s alpha (α = 0.840).
Perceived satisfaction was measured with a simple 5-point Likert
scale ranging from 1- “not at all satised” to 5 “very satised” adapted
from the Customer Satisfaction Score (CSAT) (Kiradoo, 2019).
Finally, subjective usability, as the user’s perception of how simple
and eective it is to use the learning interface (Vlachogianni and
Tselios, 2022), was measured using the System Usability Scale (SUS)
(Brooke, 1996), with ten items over three dimensions: eectiveness,
eciency, and satisfaction (ISO 9241-11). All items were evaluated on
a 1–5 Likert scale ranging from 1 “Strongly disagree” to 5 “Strongly
agree.” Scores were then converted to scores ranging from 0–100 using
the original calculation (Brooke, 1996). e internal consistency tests
revealed an acceptable Cronbach’s alpha (α = 0.798).
3.4.3 Learning gains
e participants’ answers to all questions for the learning task
were extracted aer task completion to measure the evolution of the
learning gains throughout the experiment. A score of 1 or 0 was
assigned for each correct or incorrect answer, respectively. All scores
were compiled into a single le, and participants were associated with
their performance data per block. Finally, learning gains for each
participant were calculated by subtracting the scores of block 1 from
the scores of block 2, block 3, and block 4.
3.5 Procedure
e average experimental session lasted approximately two
hours. Participants were rst greeted, provided with an explanation
of the study, and then signed consent to participate. e experiment
took place in a custom-built soundproof Faraday cage to protect
the EEG signal from external electromagnetic interference. e
experiment was monitored through a one-way mirror and shared
computer screens in an adjacent room. Participants were seated in
a chair in front of a computer screen, and a keyboard and mouse
were provided to interact with the IUI. Once seated, participants
were asked to complete the pre-test questionnaire. Shortly aer,
their head measurements were taken to t the EEG cap and
sensors, the amplier was turned on, and electroconductive gel was
applied to each electrode before impedance testing (< 7kOhm).
e BCI model was then started, and the participants’ le was
created to save their EEG data. Consequently, the researcher
selected the correct interface type (regular or adaptive) in the IUI
according to the participant’s number, which created the
participant’s le in the learning interface.
Participants began with a calibration phase, consisting of the 90-s
baseline task, followed by the n-back task to personalize cognitive load
thresholds. e calibration phase was directly followed by the learning
task, where participants rst had to read the study instructions on the
IUI’s landing page and wait for the researchers’ signal to start the task.
ey were asked to sit in a comfortable position and to limit head and
body movements. For the AMgroup, participants were informed that
their overall performance would beevaluated and that they should
aim for the highest score possible to gain more prize draw tickets.
Participants then started the task, which consisted of 4 blocks
separated by 30-s breaks. Aer the experiment, participants were
asked to complete a post-test questionnaire.
TABLE2 Questionnaires used in this study, with Cronbach’s alpha measure for multiple-item questionnaires.
Measure Questionnaires used Items Cronbach’s alpha
Before the experiment
Student status Yes or no question 1 –
Prior level of interest 10-point Likert scale 1 –
Knowledge on constellations
Short, reliable measure of subjective Knowledge
questionnaire
10 0.772
After the experiment
Self-perceived usability of
interface
System usability scale 10 0.798
Self-perceived mental workload NASA-TLX 6 0.715
Self-perceived cognitive
absorption
Cognitive absorption questionnaire 15 0.840
Self-perceived satisfaction 5-point Likert scale 1 –

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 13 frontiersin.org
4 Data analysis
All statistical analyses were performed using R Studio (version
1.4.1103) using the jamovi package (version 1.2.23) to produce
descriptive statics for cognitive load values of the learning task
(derived from blocks 2 and 3), psychometric values and learning
gains. e psych package (version 2.0.12) was used to calculate
Cronbach’s alpha for the multiple-item questionnaires.
Initial data assessment showed that the data were ordinal, wethus
opted for non-parametric statistical tests. We employed one-way
independent samples Mann–Whitney U tests to compare adaptive
measures, post-test questionnaire scores and learning gains between
each group, using the wilcox.test function of the stats package (version
3.6.3). For single-tailed hypothesis, all p values obtained were divided by
2. For two-tailed hypotheses such as those involving level of interest and
knowledge (see Section 3.4.1), measures of adaptivity and the measures
derived from the SUS post-test questionnaire (see Section 3.4.2) p-values
were not divided by 2. Finally, all eect sizes were calculated using the
wilcox_esize of the rstatix package (version 0.7.2), which returns the
rank-biserial correlation by calculating r = z/√N (Rosenthal etal., 1994).
5 Results
5.1 Descriptive results of adaptive measures
We performed a Mann–Whitney U test to validate the
eectiveness of the neuro-adaptive interface and assess whether
adaptive measures occurred in response to changes in cognitive load.
Table3 provides a summary of the results for the adaptive measures
under both adaptive conditions.
e performance of the neuroadaptive interface across both
experimental groups revealed no signicant dierence in its
eectiveness (p > 0.05), suggesting its consistent responsiveness
regardless of the presence or absence of the motivational factor.
Overall, these results conrm that the IUI functioned as intended by
adapting the speed of information provision for approximately half of
the 64 constellations presented throughout each block when high and
low cognitive load levels were detected, with comparable frequency on
average for both low and high cognitive load levels across conditions.
5.2 Prior levels of interest and knowledge
of constellations and perceived usability of
the IUI
We performed further statistical testing to verify no dierences exist
between the three groups for the independent control pretest variables,
prior level of interest, knowledge of constellations and perceived usability.
For prior level of interest, no signicant dierences were reported
between the three groups (p > 0.05), group C (Mdn = 5.00), group A
(Mdn = 5.00) and group AM(Mdn = 4.50). Similarly, for knowledge of
constellations no signicant dierences were reported between the
three groups (p > 0.05), group C (Mdn = 3.22), group A (Mdn = 3.00)
and group AM(Mdn = 3.33).
Furthermore, no signicant dierences between the three groups
were reported for the perceived usability of the interface (p > 0.05),
group C (Mdn = 80.00), group A (Mdn = 78.75) and group
AM(Mdn = 85.00). Moreover, the usability scores indicated above the
“good usability” threshold of the scale’s interpretation (Brooke, 1996),
group C (M = 82.06, SD = 8.02), group A (M = 77.95, SD = 10.98) and
group AM(M = 81.09, SD = 13.51), conrming that the perceived
usability of the interface did not inuence.
5.3 Learning gains
As indicated in Figure8, group C and AMachieve greater learning
gains than group A. Specically, the AMgroup achieved the greatest
learning gains. Statistical testing of the learning gains between groups
C and A revealed no signicant dierence (p > 0.05) (Table 4),
providing no support for our hypothesis (H1a), which states that
neuro-adaptivity leads to greater learning gains compared to the
absence of neuro-adaptivity. However, the AMgroup had signicantly
TABLE3 Descriptive analysis of adaptive measures for both adaptive groups across blocks 2 and 3.
Group A (n= 22) Group AM(n= 16) Mann–Whitney
U
M SD Mdn Max Min M SD Mdn Max Min U p
Block 2
Tot a l 33.55 8.77 34.00 52.00 16.00 32.94 7.18 33.00 47.00 21.00 179.50 0.929
Low CL 15.73 4.72 15.50 26.00 7.00 15.38 3.70 15.00 23.00 9.00 179.00 0.941
High CL 17.83 4.10 18.50 26.00 9.00 17.56 3.52 18.00 24.00 12.00 187.00 0.755
Block 3
Tot a l 34.18 8.46 34.50 48.00 19.00 33.88 6.93 35.00 47.00 23.00 177.50 0.976
Low CL 15.82 4.25 16.00 23.00 8.00 15.81 3.37 16.00 22.00 11.00 175.00 0.988
High CL 18.36 4.24 18.50 25.00 11.00 18.06 3.62 19.00 25.00 12.00 181.50 0.881
Blocks 2 and 3 combined
Tot a l 67.73 15.07 66.50 100.00 45.00 66.81 9.08 69.00 81.00 50.00 169.00 0.848
Low CL 31.55 7.94 31.00 49.00 20.00 31.19 4.55 32.00 38.00 22.00 162.50 0.700
High CL 36.18 7.20 36.00 51.00 24.00 35.63 4.63 37.00 43.00 26.00 175.00 0.988

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 14 frontiersin.org
higher learning gains than group A between block 1 and 2 (U = 109.50,
p = 0.023, r = 0.32), between block 1 and 3 (U = 88.00, p = 0.005,
r = 0.42), and between block 1 and 4 (U = 70.00, p = 0.001, r = 0.51).
Indicating that as the learning task progressed, the eect size became
stronger, suggesting a greater impact of the motivational factor on the
adaptive intervention. ese ndings support our hypothesis (H2a),
which states that adding motivation to neuro-adaptivity helps to
achieve greater learning gains compared to neuro-adaptivity alone.
5.4 Perceived mental workload
Based on the perceived cognitive workload questionnaire’s
interpretation table (Hart, 1986), groups A and AMreported a
“somewhat high” mean score of perceived mental workload, while
group C reported a “somewhat high” to “borderline high” mean
score (Table 5), indicating the highest level of mental workload.
However, no signicant dierence between groups C and A, nor
FIGURE8
Learning gains throughout the experiment for each group (* p < 0.05, ** p < 0.01).
TABLE4 Descriptive statistics and between-subjects analyses for the learning gains (* p < 0.05, ** p < 0.01).
Research question 1: comparing group C and group A (n= 39)
Group C (n= 17) Group A (n= 22) Mann–Whitney U test
M SD Mdn M SD Mdn U p r
Between Block 1
and 2 34.93 10.30 32.81 33.24 12.41 32.03 200.00 0.362 -
Between Block 1
and 3 46.78 11.27 46.88 45.88 11.60 45.31 194.50 0.423 -
Between Block 1
and 4 51.01 8.68 51.01 48.79 10.32 47.66 209.00 0.273 -
Research question 2: comparing group A and group AM(n= 38)
Group A (n= 22) Group AM(n= 16) Mann–Whitney U test
M SD Mdn M SD Mdn U p r
Between Block 1
and 2 33.24 12.41 32.03 40.72 9.45 40.63 109.50 0.023 *0.32
Between Block 1
and 3 45.88 11.60 45.31 55.76 9.42 56.25 88.00 0.005 ** 0.42
Between Block 1
and 4 48.79 10.32 47.66 60.45 8.85 60.16 70.00 0.001 ** 0.51
Learning gains range from 0–100.

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 15 frontiersin.org
between groups A and AMfor the overall questionnaire (p > 0.05)
were reported. ese ndings provide no support for our hypothesis
(H2b), which states that adding motivation to neuro-adaptivity
reduces perceived mental workload compared to neuro-adaptivity
alone. However, individual analysis for the Temporal Demand
dimension reported a signicant dierence between group A and C
(U = 268.50, p = 0.012, r = 0.37) in that, participants in group A
reported feeling signicantly less time pressure. ese ndings
partially support our hypothesis (H1b), which states that neuro-
adaptivity reduces perceived mental workload compared to the
absence of neuro-adaptivity.
5.5 Perceived cognitive absorption
No signicant dierence between groups C, A and AMwere
reported for overall perceived cognitive absorption (p > 0.05) (Table6).
However, individual analysis of the Heightened Enjoyment dimension
reported that group A reported feeling signicantly less enjoyment in
completing the learning task than group C (U = 292.5, p = 0.002,
r = 0.48). ese ndings do not support our hypothesis (H1c), which
states that neuro-adaptivity generates a higher level of self-perceived
cognitive absorption compared to the absence of neuro-adaptivity. For
the same dimension, the AMgroup reported a signicantly higher
level of enjoyment compared to the A group (U = 118.00, p = 0.044,
r = 0.28). Additionally, individual analysis of the Curiosity dimension
revealed that the AM group reported feeling signicantly more
curious about constellations and the learning interface compared to
the A group (U = 97.50, p = 0.011, r = 0.38), partially supporting our
hypothesis (H2c), which states that adding motivation to
neuro-adaptivity generates a higher level of self-perceived cognitive
absorption than neuro-adaptivity alone.
5.6 Perceived satisfaction
Groups C and AMreported a higher mean score of self-perceived
satisfaction than group A (Table7). Group C reported the highest
mean score of the three groups. Statistical testing revealed that group
A reported feeling signicantly less satised with their learning
experience compared to group C (U = 261.5, p = 0.014, r = 0.35).
However, no signicant dierence was found between groups A and
AM(p > 0.05). ese ndings do not support our hypothesis (H1d),
which states that neuro-adaptivity generates a higher level of self-
perceived satisfaction of the learning experience compared to the
absence of neuro-adaptivity. Furthermore. these ndings provide no
support for our hypothesis (H2d), which states that adding motivation
to neuro-adaptivity generates a higher level of self-perceived
satisfaction of the learning experience compared to neuro-
adaptivity alone.
6 Discussion
Our results suggest that adapting the learning speed of a
memorization-based learning task, when combined with a
motivational factor, leads to greater learning gains and greater
curiosity and enjoyment when performing the learning task. It appears
that motivation plays a role in inuencing these results, and it is
evident that it had a signicant impact on the neuro-adaptive
TABLE5 Descriptive statistics and between-subjects analyses for the perceived mental workload (* p < 0.05, ** p < 0.01).
Research question 1: comparing group C and group A (n= 39)
Group C (n= 17) Group A (n= 22) Mann–Whitney U test
M SD Mdn M SD Mdn U p r
Perceived mental
workload (0–100) 48.29 14.42 50.00 40.05 13.54 39.50 244.00 0.055 -
Mental demand 56.76 20.84 60.00 46.82 23.12 45.00 234.00 0.093 -
Temporal demand 53.53 19.26 55.00 38.18 24.71 30.00 268.50 0.012 *0.37
Performance 30.29 17.54 25.00 27.05 16.95 25.00 211.50 0.247 -
Eort 62.35 22.92 65.00 60.45 18.32 67.50 205.00 0.309 -
Frustration 38.53 23.17 40.00 27.73 23.49 20.00 242.00 0.061 -
Research question 2: comparing group A and group AM(n= 38)
Group A (n= 22) Group AM(n= 16) Mann–Whitney U test
M SD Mdn M SD Mdn U p r
Perceived mental
workload (0–100) 40.05 13.54 39.50 41.06 16.81 41.00 179.00 0.471 -
Mental demand 46.82 23.12 45.00 44.38 28.22 30.00 192.00 0.323 -
Temporal demand 38.18 24.71 30.00 44.69 26.92 40.00 152.00 0.242 -
Performance 27.05 16.95 25.00 26.56 15.68 25.00 172.50 0.465 -
Eort 60.45 18.32 67.50 55.63 21.67 60.00 205.00 0.199 -
Frustration 27.73 23.49 20.00 34.06 22.38 30.00 141.50 0.155 -

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 16 frontiersin.org
interface’s eectiveness. ese results emphasize the importance of
considering motivational strategies and interface design in developing
adaptive learning interfaces to optimize learning experiences.
First, our results suggest that motivation plays a critical role in
achieving greater learning gains. Even though participants used the
same adaptive IUI, the AMgroup outperformed the A group. is
nding could beexplained by the presence of the motivating factor,
which may have led participants to become more invested and
persistent in completing the learning task. Furthermore, this result
aligns with the current literature, suggesting that extrinsic motivation
is important in improving test results (Liu etal., 2012). Extrinsic
motivation has been suggested to cultivate motivation when beginning
learning experiences, which may develop into intrinsic motivation as
the learning process progresses (Tohidi and Jabbari, 2012). Potentially,
participants in the AM group may have been motivated by the
nancial incentive at rst, which may have grown into intrinsic
motivation due to the length of the learning task. erefore, in the
current study, the nancial incentive may have been a driving force to
complete the learning task, which led to higher levels of enjoyment
and curiosity as expected from intrinsic motivation. is conclusion
TABLE6 Descriptive statistics and between-subjects analyses for the perceived cognitive absorption (* p < 0.05, ** p < 0.01).
Research question 1: comparing group C and group A (n= 39)
Group C (n= 17) Group A (n= 22) Mann–Whitney U test
M SD Mdn M SD Mdn U p r
Perceived cognitive
absorption (1–7) 4.67 0.76 4.47 4.27 0.85 4.23 245.00 0.052 –
Temporal
dissociation 3.98 1.31 4.00 3.52 1.26 3.17 230.50 0.111 –
Focused immersion 5.12 1.25 5.00 4.94 1.53 5.17 190.00 0.472 –
Heightened
Enjoyment 4.71 1.03 4.33 3.64 1.19 3.67 292.50 0.002 ** 0.48
Curiosity 4.12 1.29 4.33 3.88 1.37 3.83 210.00 0.261 –
Control 5.43 0.89 5.67 5.36 0.72 5.33 207.50 0.284 -
Research question 2: comparing group A and group AM(n= 38)
Group A (n= 22) Group AM(n= 16) Mann–Whitney U test
M SD Mdn M SD Mdn U p r
Perceived cognitive
absorption (1–7) 4.27 0.85 4.23 4.70 0.72 4.67 120.50 0.052 –
Temporal
dissociation 3.52 1.26 3.17 3.21 1.34 2.67 200.50 0.238 –
Focused immersion 4.94 1.53 5.17 5.60 0.60 5.67 138.00 0.133 –
Heightened
enjoyment 3.64 1.19 3.67 4.35 1.43 4.67 118.00 0.044 *0.28
Curiosity 3.88 1.37 3.83 4.81 1.10 5.00 97.50 0.011 *0.38
Control 5.36 0.72 5.33 5.50 0.73 5.50 143.00 0.165 –
TABLE7 Descriptive statistics and between-subjects analyses for the perceived satisfaction (* p < 0.05, ** p < 0.01).
Research question 1: comparing group C and group A (n= 39)
Group C (n= 17) Group A (n= 22) Mann–Whitney U test
M SD Mdn M SD Mdn U p r
Perceived
satisfaction (1–5) 4.12 0.78 4.00 3.36 1.14 3.00 261.50 0.014 *0.35
Research question 2: Comparing group A and group AM(n= 38)
Group A (n= 22) Group AM(n= 16) Mann–Whitney U test
M SD Mdn M SD Mdn U p r
Perceived
satisfaction (1–5) 3.36 1.14 3.00 3.81 1.05 4.00 132.50 0.093 –

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 17 frontiersin.org
is supported by research showing that while extrinsic motivational
factors may not have as much of a long-term impact as intrinsic
motivational factors, they can lead to high levels of engagement and
commitment in the short term and better learning performances
(Tohidi and Jabbari, 2012). Furthermore, a study by Robinson and
colleagues investigated the impact of a nancial incentive on attention
and memory test performance; their results suggest that extrinsically
motivated participants performed signicantly better at both attention
and memory tests (Robinson etal., 2012). ese results support our
ndings that show a similar eect of greater learning gains from
participants in the AM group in our memory-based learning
task results.
However, contrary to our expectations, the adaptive IUI alone
did not result in greater learning gains than the regular
non-adaptive IUI. The results indicated that the adaptive IUI led
to significantly lower enjoyment and overall satisfaction levels
than the regular non-adaptive IUI. One possible explanation for
these results is that the rapid learning pace (speed of information
presentation) imposed on the control group may have served as
an indirect extrinsic motivator. In other words, the quick
information delivery speed may have been perceived as a
competition, indirectly prompting and extrinsically motivating
participants to race against the clock. Therefore, it is conceivable
that those participants who used the adaptive IUI experienced a
decrease in motivation, potentially due to the increased length of
the task and repeated instructions, compared to those using the
fast-imposed learning speed, who may have perceived and
experienced the imposed rapid pace as an indirect extrinsic
motivator. Thus, group A may have experienced increased
boredom, negatively impacting the learning experience overall.
Comparatively, as the learning experiment progressed, it
appears that regular IUI users became more accustomed to the
swift delivery pace. However, the imposed pace did not lead to
lower learning gains as expected; instead, it appears to have
enhanced the enjoyment and satisfaction of the experience due to
a possible indirect effect on the learner’s extrinsic motivation.
This finding aligns with current literature, suggesting that a
motivated learner may have higher satisfaction and pleasure levels
while completing a task (Borah, 2021). Furthermore, the
AMgroup, which coupled interface adaptivity with financial gain,
may have overlooked the length of the task and the repetitive
instructions due to increased immersion and the added extrinsic
motivational factor, which gave a purpose to pursue and finish the
learning task. Consequently, the AMgroup reported significantly
higher enjoyment and curiosity, contributing to a greater
learning experience.
Task difficulty may also have affected the classification of the
cognitive load index and the relationship between alpha and theta
activity, which may have had downstream effects on how
responsive the adaptive interface was to changes in cognitive
workload at the participant level. As mentioned in Section 2.3,
alpha desynchronization is known to result from cognitive
processing in situations of moderate to high mental workload
during memory-based learning tasks. However, in the current
study, some participants may have struggled with the task, leading
to the solicitation of additional resources from the brain to cope
with the heightened cognitive load. Past studies have shown that
during more demanding cognitive tasks, theta synchronization
may obscure alpha desynchronization in the context of cognitive
load, leading to measurement issues (Klimesch, 1999; Klimesch
etal., 1998). In other words, increased task difficulty enhances
theta synchronization, resulting in the inhibition of alpha
desynchronization within regions of the brain measured by
EEG. In a word-memorization study by Klimesch etal. (1997),
they found a connection between theta synchronization and the
encoding and retrieval of episodic information. These findings
point to a potential limitation in the design of the BCI used in this
study, given that the classification index used only considers alpha
activity at the parietal P7 electrode (see Section 6.1).
Furthermore, our findings partially support our hypothesis
that employing the adaptive IUI leads to a decreased mental
workload compared to the regular non-adaptive IUI. Even though
no significant differences were observed in the global score of the
mental workload questionnaire between groups A and C, the
Temporal Demand dimension did indicate a greater level of
temporal stress in group C. In other words, group C felt
significantly more time-restricted and felt hurried and rushed to
complete the learning task. More precisely, group C may have
found the learning task more challenging as they needed to
manage their cognitive load resources while keeping up with the
fast-imposed pace of the learning task. This result aligns with the
working memory resource depletion hypothesis, which suggests
that learning tasks requiring active use of working memory
resources may lead to temporary depletion and fatigue and can
place additional stress on the learner (Chen etal., 2018). Overall,
our results demonstrate the effectiveness of adjusting the speed of
information presentation, i.e., learning pace, to the learner’s real-
time cognitive load to reduce the perception of temporal stress of
the user.
6.1 Limitations and future work
First, our cognitive load classication index only includes alpha-
band activity at the parietal P7 electrode. is decision was made
based on the analysis of our pre-tests and conrmed by the current
literature (see Section 3.2.4). However, we acknowledge that this
classication approach has limitations since cognitive load induces
changes in brain activity within and between multiple cerebral regions,
and our memory-based learning task demands not only information
encoding and retrieval but also rapid decision-making as participants
must identify the correct constellation name when a constellation
image is presented. Decision-making requires manipulating multiple
pieces of information to make a decision, signicantly impacting
working memory capacity. Previous studies indicate that the prefrontal
cortex plays a central role in decision-making processes, specically
with alpha and theta oscillations (Bechara etal., 1998; Euston etal.,
2012). erefore, in future work, we shall analyze the functional
connectivity between parietal and prefrontal activity to measure both
cognitive load and decision-making processes in real-time, revealing
how information is processed and integrated. Changes in connectivity
indicate adjustments in cognitive load during memory-based learning
tasks (Katsuki and Constantinidis, 2012; Murray etal., 2017; Vincent
etal., 2008).
Second, our BCI model did not include an EEG signal artifact
ltering block. To minimize the occurrence of artifacts,

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 18 frontiersin.org
wemonitored electrode impedances and the EEG signal constantly
during the session. Additionally, welimited external inferences by
conducting the experiment within a Faraday cage. Weused active
EEG electrodes, including ampliers, to minimize artifacts and
signal degradation. Wereferred our signal to an electrode placed
on the earlobe for common rejection mode. Weintegrated a data
pre-processing block into the BCI model that had lters targeting
specic relevant frequency bands. Finally, we instructed the
participant to minimize body movements to ensure the validity of
our results. Furthermore, our index has the advantage of stabilizing
cognitive load classication by considering the last 60 s of
recording, thus reducing artifact impact on the classication.
Nevertheless, weacknowledge that the EEG signal quality used in
the experiment might have been aected occasionally by some
artifacts or muscle noise.
Finally, our study’s experimental design did not include a
fourth group specically tailored to investigate the impact of
motivation in the absence of adaptive measures. e decision to
include only three groups in our design was inuenced by practical
considerations, such as resource availability, and by existing
theoretical foundations. Previous studies conducted in learning
contexts without BCI technology have demonstrated that extrinsic
and intrinsic motivation signicantly impact learning gains and
performance (Gong etal., 2021; Liu etal., 2012; Xu etal., 2021).
erefore, this design choice aimed to maintain a focused
examination of the independent and interactive eects of adaptive
measures and motivation on learning gains. In other words, the
primary focus of this study was not the eect of motivation on
learning with the regular IUI, as this has already been exhaustively
studied and found to have a signicant impact. However,
werecognize that the inclusion of a fourth group of participants
who complete the learning experience using the regular IUI and
with the presence of the motivational factor could provide deeper
insight into the interaction between the adaptive measures and
motivation and could enhance the overall interpretation of
the results.
In the future, improving the classification of cognitive load
by considering brain networks instead of solely focusing on the
alpha activity of the P7 electrode and integrating Machine
Learning or Deep Learning tools into the BCI would
bebeneficial (Rabbani and Islam, 2024; Torres-García etal.,
2023). For example, Gogna etal. (2024) used a Support Vector
Machine (SVM) model to classify cognitive workload levels
based on physiological data (EEG) and subjective assessments
(NASA-TLX) during a “Spot the Difference” task. The SVM
model demonstrated impressive classification accuracy,
suggesting that it can effectively differentiate between varying
levels of cognitive workload. Such a model could betested when
applied to a learning task. Additionally, integrating more
advanced artifact cleaning methods into the online BCI model
would berelevant to ensure thorough data cleaning (Barachant
etal., 2013; Daly etal., 2014; Urigüen and Garcia-Zapirain,
2015). These improvements would lead to more efficient and
granular cognitive load classification by considering different
brain regions and frequency bands free from artifacts. Including
a secondary physiological measure for classifying cognitive load
or evaluating system performance, such as pupillometry data,
would be valuable. This addition would yield a more
comprehensive assessment of cognitive load and the impacts of
the system on learning experiences and outcomes. In practice,
it would also be interesting to evaluate this system among
student populations with academic challenges, such as those
with neurodevelopmental disorders like attention deficit
disorder (with or without hyperactivity). Such a system could
bea game-changer for learners who face academic challenges,
as it would enable adaptive learning that caters to their abilities.
7 Conclusion
We designed this study to investigate the impact of a neuro-
adaptive interface on the enhancement of the learning experience
using a constellation memorization-based learning task. Our aim
was to determine if a passive BCI, which adjusts the speed of
presenting information to learners based on their real-time
cognitive load levels, would enhance their learning experience by
keeping them within their ZPD. Additionally, weexplored to what
extent motivation was a prerequisite for effective adaptation. Our
study employed a between-subjects design. Participants were
assigned to either the control group, adaptive without motivation
group, or adaptive with motivation group based on their order of
enrollment in the study. Before the experiment, all participants
completed a pre-test questionnaire and the n-back task to calibrate
personal cognitive load thresholds. These thresholds were
subsequently utilized in only the two adaptive groups. In line with
previous research, wehypothesized that neuroadaptivity creates
an optimal learning environment by enhancing learning gains,
reducing self-perceived cognitive workload, generating higher
levels of self-perceived cognitive absorption, and generating a
higher level of satisfaction about the learning experience. Finally,
we expected that motivation moderates the effect of
neuroadaptation by augmenting its effectiveness and self-
perception of an optimal learning environment. To test these
hypotheses, weconducted one-way, non-parametric between-
group analyses. Our results suggest that coupling motivation and
adaptive IUI enhances learning gains for a memory-based
learning task and contributes to enhancing the overall learning
experience. However, we found no significant impact of the
adaptive IUI alone in enhancing the learning experience.
Nevertheless, we discovered that the imposed learning pace
induced a significant temporal stress perception but significantly
decreased the satisfaction level of the BCI. Our results suggest the
importance of considering motivational strategies and interface
design in developing adaptive learning interfaces to optimize
learning experiences.
By using motivation as a catalyst, our system makes it possible
to significantly improve learning gains while respecting the
individual abilities of each learner. In theory, our system addresses
the problem of lack of individual consideration and
personalization of learning according to each learner. To our
knowledge, few studies have explored the use of passive BCI
systems in educational settings. Our study contributes to
advancing knowledge by establishing a foundation for the
application of such a system in learning. In practice, our study
demonstrates the potential and feasibility of utilizing both
motivation and passive BCI to improve learning outcomes and

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 19 frontiersin.org
enhance the overall learning experience. Overall, our findings
support the pursuit of such an opportunity.
Data availability statement
e raw data supporting the conclusions of this article will
bemade available by the authors, without undue reservation.
Ethics statement
e studies involving humans were approved by HEC Research
Ethics Board (REB) (certicate 2023-5071). e studies were
conducted in accordance with the local legislation and institutional
requirements. e participants provided their written informed
consent to participate in this study.
Author contributions
NB: Conceptualization, Data curation, Formal analysis,
Investigation, Methodology, Project administration, Soware,
Validation, Visualization, Writing – original dra, Writing – review &
editing. PC: Conceptualization, Resources, Supervision, Writing –
review & editing. AK: Conceptualization, Investigation, Methodology,
Resources, Soware, Supervision, Validation, Writing – review &
editing, Writing – original dra. JB: Conceptualization, Investigation,
Methodology, Resources, Soware, Supervision, Validation, Writing
– review & editing. BT: Conceptualization, Investigation,
Methodology, Writing – review & editing. SS: Conceptualization,
Funding acquisition, Resources, Supervision, Writing – review &
editing. P-ML: Conceptualization, Funding acquisition, Resources,
Supervision, Writing – review & editing.
Funding
e author(s) declare nancial support was received for the
research, authorship, and/or publication of this article. is study was
supported by IVADO (PRF-2021-05) and NSERC
(RGPIN-2020-06048).
Acknowledgments
We are grateful to IVADO and NSERC for funding our project.
Conflict of interest
e authors declare that the research was conducted in the
absence of any commercial or nancial relationships that could
beconstrued as a potential conict of interest.
Publisher’s note
All claims expressed in this article are solely those of the authors
and do not necessarily represent those of their aliated
organizations, or those of the publisher, the editors and the
reviewers. Any product that may beevaluated in this article, or claim
that may bemade by its manufacturer, is not guaranteed or endorsed
by the publisher.
References
Abiri, R., Borhani, S., Sellers, E. W., Jiang, Y., and Zhao, X. (2019). A comprehensive
review of EEG-based brain–computer interface paradigms. J. Neural Eng. 16:011001.
doi: 10.1088/1741-2552/aaf12e
Agarwal, R., and Karahanna, E. (2000). Time ies when you're having fun: cognitive
absorption and beliefs about information technology usage. MIS Q. 24, 665–694. doi:
10.2307/3250951
Ahn, M., Lee, M., Choi, J., and Jun, S. C. (2014). A review of brain-computer Interface
games and an opinion survey from researchers, Developers and Users. Sensors 14,
14601–14633. doi: 10.3390/s140814601
Alamri, H. A., Watson, S., and Watson, W. (2021). Learning technology models that
support personalization within blended learning environments in higher education.
Tec hTr e nds 65, 62–78. doi: 10.1007/s11528-020-00530-3
Anderson, B. A. (2016). e attention habit: how reward learning shapes attentional
selection. Ann. N. Y. Acad. Sci. 1369, 24–39. doi: 10.1111/nyas.12957
Anderson, E. W., Potter, K. C., Matzen, L. E., Shepherd, J. F., Preston, G. A., and
Silva, C. T. (2011). A user study of visualization eectiveness using EEG and cognitive
load. Computer graphics forum, vol. 30. Oxford, UK: Blackwell Publishing Ltd.,
791–800.
Andreessen, L. M., Gerjets, P., Meurers, D., and Zander, T. O. (2021). Toward
neuroadaptive support technologies for improving digital reading: a passive BCI-based
assessment of mental workload imposed by text diculty and presentation speed during
reading. User Model. User-Adap. Inter. 31, 75–104. doi: 10.1007/s11257-020-09273-5
Angrisani, L., Arpaia, P., Esposito, A., Gargiulo, L., Natalizio, A., Mastrati, G., et al.
(2021). Passive and active brain-computer interfaces for rehabilitation in health 4.0.
Measurement 18:100246. doi: 10.1016/j.measen.2021.100246
Anmarkrud, Ø., Andresen, A., and Bråten, I. (2019). Cognitive load and working
memory in multimedia learning: conceptual and measurement issues. Educ. Psychol. 54,
61–83. doi: 10.1080/00461520.2018.1554484
Antonenko, P., Paas, F., Grabner, R., and van Gog, T. (2010). Using
electroencephalography to measure cognitive load. Educ. Psychol. Rev. 22, 425–438. doi:
10.1007/s10648-010-9130-y
Apicella, A., Arpaia, P., Frosolone, M., Improta, G., Moccaldi, N., and Pollastro, A.
(2022). EEG-based measurement system for monitoring student engagement in learning
4.0. Sci. Rep. 12:5857. doi: 10.1038/s41598-022-09578-y
Aricò, P., Borghini, G., Di Flumeri, G., Colosimo, A., Bonelli, S., Golfetti, A., et al.
(2016). Adaptive automation triggered by EEG-based mental workload index: a passive
brain-computer interface application in realistic air trac control environment. Front.
Hum. Neurosci. 10:539. doi: 10.3389/fnhum.2016.00539
Ayres, P. (2006). Using subjective measures to detect variations of intrinsic cognitive
load within problems. Learn. Instr. 16, 389–400. doi: 10.1016/j.learninstruc.2006.09.001
Barachant, A., Andreev, A., and Congedo, M. (2013). e Riemannian potato: An
automatic and adaptive artifact detection method for online experiments using
Riemannian geometry. TOBI workshop lV, (pp.19–20).
Barki, H., Pare, G., and Sicotte, C. (2008). Linking IT implementation and acceptance
via the construct of psychological ownership of information technology. J. Inf. Technol.
23, 269–280. doi: 10.1057/jit.2008.12
Bawa, P. (2016). Retention in online courses: exploring issues and solutions—a
literature review. SAGE Open 6:2158244015621777. doi: 10.1177/2158244015621777
Bechara, A., Damasio, H., Tranel, D., and Anderson, S. W. (1998). Dissociation of
working memory from decision making within the human prefrontal cortex. J. Neurosci.
18, 428–437. doi: 10.1523/JNEUROSCI.18-01-00428.1998
Beck, J. E. (2005). Engagement tracing: using response times to model student
disengagement. Artif. Intell. Educ. 125:88.
Birbaumer, N., Murguialday, A. R., Weber, C., and Montoya, P. (2009). Neurofeedback
and brain–computer interface: clinical applications. Int. Rev. Neurobiol. 86, 107–117. doi:
10.1016/S0074-7742(09)86008-X

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 20 frontiersin.org
Bonaci, T., Calo, R., and Chizeck, H. J. (2015). App Stores for the Brain: privacy and
security in brain-computer interfaces. IEEE Technol. Soc. Mag. 34, 32–39. doi: 10.1109/
MTS.2015.2425551
Borah, M. (2021). Motivation in learning. J. Crit. Rev. 8, 550–552.
Brooke, J. (1996). SUS-A quick and dirty usability scale. Usab. Eval. Indus. 189, 4–7.
Brouwer, A.-M., Hogervorst, M. A., van Erp, J. B. F., Heelaar, T., Zimmerman, P. H.,
and Oostenveld, R. (2012). Estimating workload using EEG spectral power and ERPs in
the n-back task. J. Neural Eng. 9:045008. doi: 10.1088/1741-2560/9/4/045008
Brunken, R., Plass, J. L., and Leutner, D. (2003). Direct measurement of cognitive load
in multimedia learning. Educ. Psychol. 38, 53–61. doi: 10.1207/S15326985EP3801_7
Butler, A. C., Karpicke, J. D., and Roediger Iii, H. L. (2008). Correcting a metacognitive
error: feedback increases retention of low-condence correct responses. J. Exp. Psychol.
Learn. Mem. Cogn. 34, 918–928. doi: 10.1037/0278-7393.34.4.918
Butteld, A., Ferrez, P. W., and Millan, J. R. (2006). Towards a robust BCI: error
potentials and online learning. IEEE Trans. Neural Syst. Rehabil. Eng. 14, 164–168. doi:
10.1109/TNSRE.2006.875555
Caplan, B., and Mendoza, J. E. (2011). “Edinburgh handedness inventory” in
Encyclopedia of clinical neuropsychology. eds. J. S. Kreutzer, J. DeLuca and B. Caplan
(New York: Springer), 928.
Carabalona, R., Grossi, F., Tessadri, A., Castiglioni, P., C aracciolo, A., and de Munari, I.
(2012). Light on! Real world evaluation of a P300-based brain–computer interface (BCI)
for environment control in a smart home. Ergonomics 55, 552–563. doi:
10.1080/00140139.2012.661083
Cavanagh, J. F., and Frank, M. J. (2014). Frontal theta as a mechanism for cognitive
control. Trends Cogn. Sci. 18, 414–421. doi: 10.1016/j.tics.2014.04.012
Chaiklin, S. (2003). e zone of proximal development in Vygotsky’s analysis of
learning and instruction. Vygotsky’s Educ. eor. Cult. Cont. 1, 39–64. doi: 10.1017/
CBO9780511840975.004
Chaudhary, U., Birbaumer, N., and Ramos-Murguialday, A. (2016). Brain–computer
interfaces for communication and rehabilitation. Nat. Rev. Neurol. 12, 513–525. doi:
10.1038/nrneurol.2016.113
Chen, O., Castro-Alonso, J. C., Paas, F., and Sweller, J. (2018). Extending cognitive load
theory to incorporate working memory resource depletion: evidence from the spacing
eect. Educ. Psychol. Rev. 30, 483–501. doi: 10.1007/s10648-017-9426-2
Chen, J., Zhang, D., Engel, A. K., Gong, Q., and Maye, A. (2017). Application of a
single-icker online SSVEP BCI for spatial navigation. PLoS One 12:e0178385. doi:
10.1371/journal.pone.0178385
Chipchase, L., Davidson, M., Blackstock, F., Bye, R., Clothier, P., Klupp, N., et al.
(2017). Conceptualising and measuring student disengagement in higher education: a
synthesis of the literature. Int. J. High. Educ. 6, 31–42. doi: 10.5430/ijhe.v6n2p31
Cohen, M. X. (2017). Where does EEG come from and what does it mean? Trends
Neurosci. 40, 208–218. doi: 10.1016/j.tins.2017.02.004
Colligan, L., Potts, H. W., Finn, C. T., and Sinkin, R. A. (2015). Cognitive workload
changes for nurses transitioning from a legacy system with paper documentation to a
commercial electronic health record. Int. J. Med. Inform. 84, 469–476. doi: 10.1016/j.
ijmedinf.2015.03.003
Csikszentmihalyi, M. (1975). Beyond boredom and anxiety. San Francisco, CA, US:
Jossey-Bass.
Csikszentmihalyi, M. (2014). “Toward a Psychology of Optimal Experience,” in Flow
and the Foundations of Positive Psychology. ed. M. Csikszentmihalyi. (Dordrecht:
Springer).
Curum, B., and Khedo, K. K. (2021). Cognitive load management in mobile learning
systems: principles and theories. J. Comput. Educ. 8, 109–136. doi: 10.1007/
s40692-020-00173-6
Daly, I., Scherer, R ., Billinger, M., and Müller-Putz, G. (2014). FORCe: fully online and
automated artifact removal for brain-computer interfacing. IEEE Trans. Neural Syst.
Rehabil. Eng. 23, 725–736. doi: 10.1109/TNSRE.2014.2346621
Daly, J. J., and Wolpaw, J. R. (2008). Brain–computer interfaces in neurological
rehabilitation. Lancet Neurol. 7, 1032–1043. doi: 10.1016/S1474-4422(08)70223-0
De Jong, T. (2010). Cognitive load theory, educational research, and instructional
design: some food for thought. Instr. Sci. 38, 105–134. doi: 10.1007/
s11251-009-9110-0
Debue, N., and van de Leemput, C. (2014). What does germane load mean? An
empirical contribution to the cognitive load theory. Front. Psychol. 5:1099. doi: 10.3389/
fpsyg.2014.01099
Dehais, F., Ladouce, S., Darmet, L., Nong, T.-V., Ferraro, G., Torre Tresols, J., et al.
(2022). Dual passive reactive brain-computer Interface: a novel approach to human-
machine Symbiosis. Front. Neuroergon. 3:824780. doi: 10.3389/fnrgo.2022.824780
DeLeeuw, K. E., and Mayer, R. E. (2008). A comparison of three measures of cognitive
load: evidence for separable measures of intrinsic, extraneous, and germane load. J.
Educ. Psychol. 100, 223–234. doi: 10.1037/0022-0663.100.1.223
Demazure, T., Karran, A., Labonté-LeMoyne, É., Léger, P.-M., S énécal, S., Fredette, M.,
et al. (2019). Sustained attention in a monitoring task: Towards a Neuroadaptative
Enterprise system Interface. Cham: Information Systems and Neuroscience.
Demazure, T., Karran, A., Léger, P.-M., Labonté-LeMoyne, É., Sénécal, S., Fredette, M.,
et al. (2021). Enhancing sustained attention. Bus. Inf. Syst. Eng. 63, 653–668. doi:
10.1007/s12599-021-00701-3
Dimoka, A., Davis, F. D., Gupta, A., Pavlou, P. A., Banker, R. D., Dennis, A. R., et al.
(2012). On the use of neurophysiological tools in IS research: developing a research
agenda for NeuroIS. MIS Q. 36, 679–702. doi: 10.2307/41703475
Dumford, A. D., and Miller, A. L. (2018). Online learning in higher education:
exploring advantages and disadvantages for engagement. J. Comput. High. Educ. 30,
452–465. doi: 10.1007/s12528-018-9179-z
Eldenfria, A., and Al-Samarraie, H. (2019). Towards an online continuous adaptation
mechanism (OCAM) for enhanced engagement: an EEG study. Int. J. Hum. Comput.
Intera ct. 35, 1960–1974. doi: 10.1080/10447318.2019.1595303
El-Sabagh, H. A. (2021). Adaptive e-learning environment based on learning styles
and its impact on development students' engagement. Int. J. Educ. Technol. High. Educ.
18, 1–24. doi: 10.1186/s41239-021-00289-4
Euston, D. R., Gruber, A. J., and McNaughton, B. L. (2012). e role of medial
prefrontal cortex in memory and decision making. Neuron 76, 1057–1070. doi:
10.1016/j.neuron.2012.12.002
Ferrer, J., Ringer, A., Saville, K., Parris, A., and Kashi, K. (2022). Students’ motivation
and engagement in higher education: the importance of attitude to online learning. High.
Educ. 83, 317–338. doi: 10.1007/s10734-020-00657-5
Festinger, D. S., Marlowe, D. B., Cro, J. R., Dugosh, K. L., Arabia, P. L., and
Benasutti, K. M. (2009). Monetary incentives improve recall of research consent
information: it pays to remember. Exp. Clin. Psychopharmacol. 17, 99–104. doi: 10.1037/
a0015421
Fini, A. (2009). e technological dimension of a massive open online course: the case
of the CCK08 course tools. Int. Rev. Res. Open Distrib. Learn. 10, 2–26. doi: 10.19173/
irrodl.v10i5.643
Flynn, L. R., and Goldsmith, R. E. (1999). A short, reliable measure of subjective
knowledge. J. Bus. Res. 46, 57–66. doi: 10.1016/S0148-2963(98)00057-5
Galy, E., Paxion, J., and Berthelon, C. (2018). Measuring mental workload with the
NASA-TLX needs to examine each dimension rather than relying on the global score:
an example with driving. Ergonomics 61, 517–527. doi:
10.1080/00140139.2017.1369583
Gao, X., Wang, Y., Chen, X., and Gao, S. (2021). Interface, interaction, and intelligence
in generalized brain–computer interfaces. Trends Cogn. Sci. 25, 671–684. doi: 10.1016/j.
tics.2021.04.003
Gerjets, P., Walter, C., Rosenstiel, W., Bogdan, M., and Zander, T. O. (2014). C ognitive
state monitoring and the design of adaptive instruction in digital environments: lessons
learned from cognitive workload assessment using a passive brain-computer interface
approach. Front. Neurosci. 8:385. doi: 10.3389/fnins.2014.00385
Gevins, A., and Smith, M. E. (2003). Neurophysiological measures of cognitive
workload during human-computer interaction. eor. Issues Ergon. Sci. 4, 113–131. doi:
10.1080/14639220210159717
Ghafurian, M., Reitter, D., and Ritter, F. E. (2020). Countdown timer speed: a trade-o
between delay duration perception and recall. ACM Transact. Comput. Hum. Interact.
27, 1–25. doi: 10.1145/3380961
Gogna, Y., Tiwari, S., and Singla, R. (2024). Evaluating the performance of the
cognitive workload model with subjective endorsement in addition to EEG. Med. Biol.
Eng. Comput. 62, 2019–2036. doi: 10.1007/s11517-024-03049-4
Gong, J., Liu, T. X., and Tang, J. (2021). How monetary incentives improve outcomes
in MOOCs: evidence from a eld experiment. J. Econ. Behav. Organ. 190, 905–921. doi:
10.1016/j.jebo.2021.06.029
Grimes, D., Tan, D. S., Hudson, S. E., Shenoy, P., and Rao, R. P. (2008). Feasibility and
pragmatics of classifying working memory load with an electroencephalograph.
Proceedings of the SIGCHI conference on human factors in computing systems,
(pp.835–844).
Grizioti, M., and Kynigos, C. (2020). “Computer-based learning, computational
thinking, and constructionist approaches” in Encyclopedia of education and information
technologies.ed. A. Tatnall (Switzerland AG: Springer), 355–371.
Gu, X., Cao, Z., Jolfaei, A., Xu, P., Wu, D., Jung, T. P., et al. (2021). EEG-based
brain-computer interfaces (BCIs): a survey of recent studies on signal sensing
technologies and computational intelligence approaches and their applications.
IEEE/ACM Trans. Comput. Biol. Bioinform. 18, 1645–1666. doi: 10.1109/
TCBB.2021.3052811
Hancock, P. A., and Meshkati, N. (1988). Human mental workload. Amsterdam
North-Holland: Elsevier Science Publishers B.V.
Hart, S. G. (1986). NASA task load index (TLX). (NASA) Ames Research Center
Moett Field, CA United States.
Hart, S. G. (2006). NASA-task load index (NASA-TLX); 20 years later. Proceedings of
the Human Factors and Ergonomics Society Annual Meeting, 50 (904). 904, 908
Hart, S. G., and Staveland, L. E. (1988). “Development of NASA-TLX (task load
index): results of empirical and theoretical research” in Advances in psychology, vol. 52
eds. P. A. Hancock and N. Meshkati (Amsterdam North-Holland: Elsevier Science
Publishers B.V.), 139–183.

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 21 frontiersin.org
Hasler, B. S., Kersten, B., and Sweller, J. (2007). Learner control, cognitive load and
instructional animation. Appl. Cogn. Psychol. 21, 713–729. doi: 10.1002/acp.1345
Hedegaard, M. (2012). “e zone of proximal development as basis for instruction”
in An introduction to Vygotsky ed. H. Daniels (Cambridge, United Kingdom:
Cambridge University Press), 234–258.
Hogervorst, M. A., Brouwer, A. M., and van Erp, J. B. (2014). Combining and
comparing EEG, peripheral physiology and eye-related measures for the assessment of
mental workload. Front. Neurosci. 8:322. doi: 10.3389/fnins.2014.00322
Hu, A., Shewokis, P. A., Ting, K., and Fung, K. (2016). Motivation in computer-
assisted instruction. Laryngoscope 126, S5–S13. doi: 10.1002/lary.26040
Kalyuga, S., and Liu, T.-C. (2015). Guest editorial: managing cognitive load in
technology-based learning environments. J. Educ. Technol. Soc. 18, 1–8.
Karran, A. J., Demazure, T., Leger, P.-M., Labonte-LeMoyne, E., Senecal, S.,
Fredette, M., et al. (2019). Toward a hybrid passive BCI for the modulation of sustained
attention using EEG and fNIRS. Front. Hum. Neurosci. 13:393. doi: 10.3389/
fnhum.2019.00393
Katsuki, F., and Constantinidis, C. (2012). Unique and shared roles of the posterior
parietal and dorsolateral prefrontal cortex in cognitive functions. Front. Integr. Neurosci.
6:17. doi: 10.3389/fnint.2012.00017
Kerous, B., Skola, F., and Liarokapis, F. (2018). EEG-based BCI and video games: a
progress report. Virtual Reality 22, 119–135. doi: 10.1007/s10055-017-0328-x
Kiradoo, G. (2019). Soware engineering quality to enhance the customer satisfaction
level of the organization. Int. J. Adv. Res. Eng. Technol. 10, 297–302.
Klašnja-Milićević, A., Vesin, B., Ivanović, M., and Budimac, Z. (2011). E-learning
personalization based on hybrid recommendation strategy and learning style
identication. Comput. Educ. 56, 885–899. doi: 10.1016/j.compedu.2010.11.001
Klimesch, W. (1999). EEG alpha and theta oscillations reect cognitive and memory
performance: a review and analysis. Brain Res. Rev. 29, 169–195. doi: 10.1016/
S0165-0173(98)00056-3
Klimesch, W., D oppelmayr, M., Russegger, H., Pachinger, T., and Schwaiger, J. (1998).
Induced alpha band power changes in the human EEG and attention. Neurosci. Lett. 244,
73–76. doi: 10.1016/S0304-3940(98)00122-0
Klimesch, W., Doppelmayr, M., Schimke, H., and Ripper, B. (1997). eta
synchronization and alpha desynchronization in a memory task. Psychophysiology 34,
169–176. doi: 10.1111/j.1469-8986.1997.tb02128.x
Klimesch, W., Schack, B., and Sauseng, P. (2005). e functional signicance of
theta and upper alpha oscillations. Exp. Psychol. 52, 99–108. doi:
10.1027/1618-3169.52.2.99
Kosmyna, N., and Maes, P. (2019). AttentivU: an EEG-based closed-loop biofeedback
system for real-time monitoring and improvement of engagement for personalized
learning. Sensors 19:5200. doi: 10.3390/s19235200
Kosmyna, N., Tarpin-Bernard, F., Bonnefond, N., and Rivet, B. (2016). Feasibility of
BCI control in a realistic smart home environment. Front. Hum. Neurosci. 10:416. doi:
10.3389/fnhum.2016.00416
Krol, L. R., and Zander, T. O. (2017). Passive BCI-based Neuroadaptive systems.
Proceedings of the 7th Graz brain-computer Interface conference,
Ku, H.-Y., and Sullivan, H. J. (2002). Student performance and attitudes using
personalized mathematics instruction. Educ. Technol. Res. Dev. 50, 21–34. doi: 10.1007/
BF02504959
Kulhavy, R. W. (1977). Feedback in written instruction. Rev. Educ. Res. 47, 211–232.
doi: 10.3102/00346543047002211
Laar, B. V. D., Gürkök, H., Bos, D. P.-O., Poel, M., and Nijholt, A. (2013). Experiencing
BCI control in a popular computer game. IEEE Transactions on Computational
Intelligence and AI in Games, 5(2), 176–184.
Labonte-Lemoyne, E., Courtemanche, F., Louis, V., Fredette, M., Sénécal, S., and
Léger, P.-M. (2018). Dynamic threshold selection for a biocybernetic loop in an adaptive
video game context. Front. Hum. Neurosci. 12:282. doi: 10.3389/fnhum.2018.00282
Lalor, E. C., Kelly, S. P., Finucane, C., Burke, R., Smith, R., Reilly, R. B., et al.
(2005). Steady-state VEP-based brain-computer Interface control in an immersive
3D gaming environment. EURASIP J. Adv. Signal Proc. 2005:706906. doi: 10.1155/
ASP.2005.3156
Lécuyer, A., Lotte, F., Reilly, R. B., Leeb, R., Hirose, M., and Slater, M. (2008). Brain-
computer interfaces, virtual reality, and videogames. Computer 41, 66–72. doi: 10.1109/
MC.2008.410
Lepper, M. R., and Malone, T. W. (2021). “Intrinsic motivation and instructional
eectiveness in computer-based education,” in Aptitude, learning, and instruction. eds.
R. E. Snow and M. J. Farr (London, United Kingdom: Routledge), 255–286.
Liang, S. F., Shaw, F. Z., Young, C. P., Chang, D. W., and Liao, Y. C. (2010). A closed-
loop brain computer interface for real-time seizure detection and control. 2010 annual
international conference of the IEEE engineering in medicine and biology,
(pp.4950–4953). IEEE.
Lin, C. T., Lin, B. S., Lin, F. C., and Chang, C. J. (2014). Brain computer Interface-based
smart living environmental auto-adjustment control system in UPnP home networking.
IEEE Syst. J. 8, 363–370. doi: 10.1109/JSYST.2012.2192756
Liu, O. L., Bridgeman, B., and Adler, R. M. (2012). Measuring learning outcomes in
higher education: motivation matters. Educ. Res. 41, 352–362. doi:
10.3102/0013189X12459679
Lotte, F., Bougrain, L., Cichocki, A., Clerc, M., Congedo, M.,
Rakotomamonjy, A., et al. (2018). A review of classification algorithms for EEG-
based brain–computer interfaces: a 10 year update. J. Neural Eng. 15:31005. doi:
10.1088/1741-2552/aab2f2
Maksimenko, V. A., Van Heukelum, S., Makarov, V. V., Kelderhuis, J., Lüttjohann, A.,
Koronovskii, A. A., et al. (2017). Absence seizure control by a brain computer interface.
Sci. Rep. 7:2487. doi: 10.1038/s41598-017-02626-y
Mamolo, L. A. (2022). Online learning and students’ mathematics motivation, self-
ecacy, and anxiety in the “new Normal”. Educ. Res. Int. 2022, 1–10. doi:
10.1155/2022/9439634
Marchesi, M., and Riccò, B. (2013). BRAVO: a brain virtual operator for education
exploiting brain-computer interfaces. In CHI'13 extended abstracts on human factors
in computing systems (pp.3091–3094).
Martin, F., and Bolliger, D. U. (2022). Developing an online learner satisfaction
framework in higher education through a systematic review of research. Int. J. Educ.
Technol. High. Educ. 19, 1–21. doi: 10.1186/s41239-022-00355-5
Mashrur, F. R., Rahman, K. M., Miya, M. T. I., Vaidyanathan, R., Anwar, S. F.,
Sarker, F., et al. (2022). BCI-based Consumers' choice prediction from EEG signals: an
intelligent Neuromarketing framework. Front. Hum. Neurosci. 16:861270. doi: 10.3389/
fnhum.2022.861270
Mertens, U., Finn, B., and Lindner, M. A. (2022). Eects of computer-based feedback
on lower-and higher-order learning outcomes: a network meta-analysis. J. Educ. Psychol.
114, 1743–1772. doi: 10.1037/edu0000764
Murray, J. D., Jaramillo, J., and Wang, X.-J. (2017). Working memory and decision-
making in a frontoparietal circuit model. J. Neurosci. 37, 12167–12186. doi: 10.1523/
JNEUROSCI.0343-17.2017
Mutlu-Bayraktar, D., Cosgun, V., and Altan, T. (2019). Cognitive load in multimedia
learning environments: a systematic review. Comput. Educ. 141:103618. doi: 10.1016/j.
compedu.2019.103618
Najjar, L. J. (1996). Multimedia information and learning. J. Educ. Multimed.
Hypermedia 5, 129–150.
Nikou, S. A., and Economides, A. A. (2016). The impact of paper-based,
computer-based and mobile-based self-assessment on students' science motivation
and achievement. Comput. Hum. Behav. 55, 1241–1248. doi: 10.1016/j.
chb.2015.09.025
O'Byrne, W. I., and Pytash, K. E. (2015). Hybrid and blended learning: modifying
pedagogy across path, pace, time, and place. J. Adolesc. Adult. Lit. 59, 137–140. doi:
10.1002/jaal.463
Paas, F., Renkl, A., and Sweller, J. (2004). Cognitive load theory: instructional
implications of the interaction between information structures and cognitive
architecture. Instr. Sci. 32, 1–8. doi: 10.1023/B:TRUC.0000021806.17516.d0
Paas, F., Tuovinen, J. E., Van Merrienboer, J. J., and Aubteen Darabi, A. (2005). A
motivational perspective on the relation between mental eort and performance:
optimizing learner involvement in instruction. Educ. Technol. Res. Dev. 53, 25–34. doi:
10.1007/BF02504795
Padeld, N., Zabalza, J., Zhao, H., Masero, V., and Ren, J. (2019). EEG-based brain-
computer interfaces using motor-imagery: techniques and challenges. Sensors 19:1423.
doi: 10.3390/s19061423
Papalambros, N. A., Santostasi, G., Malkani, R. G., Braun, R., Weintraub, S.,
Paller, K. A., et al. (2017). Acoustic enhancement of sleep slow oscillations and
concomitant memory improvement in older adults. Front. Hum. Neurosci. 11:109. doi:
10.3389/fnhum.2017.00109
Pels, E. G. M., Aarnoutse, E. J., Leinders, S., Freudenburg, Z. V., Branco, M. P., van der
Vijgh, B. H., et al. (2019). Stability of a chronic implanted brain-computer interface in
late-stage amyotrophic lateral sclerosis. Clin. Neurophysiol. 130, 1798–1803. doi:
10.1016/j.clinph.2019.07.020
Petko, D., Schmid, R., and Cantieni, A. (2020). Pacing in serious games: exploring the
eects of presentation speed on cognitive load, engagement and learning gains. Simul.
Gaming 51, 258–279. doi: 10.1177/1046878120902502
Phan, H., Andreotti, F., Cooray, N., Chén, O. Y., and Vos, M. D. (2019). Joint
classification and prediction CNN framework for automatic sleep stage
classification. IEEE Trans. Biomed. Eng. 66, 1285–1296. doi: 10.1109/
TBME.2018.2872652
Pope, A. T., B ogart, E. H., and Bartolome, D. S. (1995). Biocybernetic system evaluates
indices of operator engagement in automated task. Biol. Psychol. 40, 187–195. doi:
10.1016/0301-0511(95)05116-3
Rabbani, M. H. R., and Islam, S. M. R. (2024). Deep learning networks based decision
fusion model of EEG and fNIRS for classication of cognitive tasks. Cogn. Neurodyn.
18, 1489–1506. doi: 10.1007/s11571-023-09986-4
Riedl, R., Davis, F. D., and Hevner, A. (2014). Towards a NeuroIS research
methodology: intensifying the discussion on methods, tools, and measurement. J. Assoc.
Inf. Syst. 15, I–XXXV. doi: 10.17705/1jais.00377

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 22 frontiersin.org
Riedl, R., and Léger, P.-M. (2016). “Tools in NeuroIS research: an overview” in
Fundamentals of NeuroIS: Information systems and the brain. eds. R. Riedl and P. Léger
(Berlin Heidelberg: Springer), 47–72.
Riopel, M., Chastenay, P., Fortin-Clément, G., Potvin, P., Masson, S., and Charland, P.
(2017). Using invariance to model practice, forgetting, and spacing eects. Edulearn17
proceedings, (pp.4334–4341). IATED.
Robinson, L. J., Stevens, L. H., reapleton, C. J., Vainiute, J., McAllister-Williams, R. H.,
and Gallagher, P. (2012). Eects of intrinsic and extrinsic motivation on attention and
memory. Acta Psychol. 141, 243–249. doi: 10.1016/j.actpsy.2012.05.012
Rosenthal, R., Cooper, H., and Hedges, L. (1994). Parametric measures of eect size.
e handbook of research synthesis. eds. H. Cooper and L. Hedges (Russell Sage
Foundation) 621, 231–244.
Rosso, O. A., Blanco, S., Yordanova, J., Kolev, V., Figliola, A., Schürmann, M., et al.
(2001). Wavelet entropy: a new tool for analysis of short duration brain electrical signals.
J. Neurosci. Methods 105, 65–75. doi: 10.1016/S0165-0270(00)00356-3
Rousu, M. C., Corrigan, J. R., Harris, D., Hayter, J. K., Houser, S., Lafrancois, B. A.,
et al. (2015). Do monetary incentives matter in classroom experiments? Eects on
course performance. J. Econ. Educ. 46, 341–349. doi: 10.1080/00220485.2015.1071214
Roy, R. N., Bonnet, S., Charbonnier, S., and Campagne, A. (2013). Mental fatigue and
working memory load estimation: Interaction and implications for EEG-based passive
BCI. 2013 35th annual international conference of the IEEE engineering in medicine
and biology society (EMBC), (pp.6607–6610). IEEE.
Ryan, R. M., and Deci, E. L. (2000a). Intrinsic and extrinsic motivations: classic
denitions and new directions. Contemp. Educ. Psychol. 25, 54–67. doi: 10.1006/
ceps.1999.1020
Ryan, R. M., and Deci, E. L. (2000b). Self-determination theory and the facilitation of
intrinsic motivation, social development, and well-being. Am. Psychol. 55, 68–78. doi:
10.1037/0003-066X.55.1.68
Saadé, R., and Bahli, B. (2005). e impact of cognitive absorption on perceived
usefulness and perceived ease of use in on-line learning: an extension of the technology
acceptance model. Inf. Manag. 42, 317–327. doi: 10.1016/j.im.2003.12.013
Salimon, M. G., Sanuri, S. M. M., Aliyu, O. A., Perumal, S., and Yusr, M. M. (2021).
E-learning satisfaction and retention: a concurrent perspective of cognitive absorption,
perceived social presence and technology acceptance model. J. Syst. Inf. Technol. 23,
109–129. doi: 10.1108/JSIT-02-2020-0029
Schildberg-Hörisch, H., and Wagner, V. (2020). Monetary and non-monetary
incentives for educational attainment: design and eectiveness. Econ. Educ., 249–268.
doi: 10.1016/B978-0-12-815391-8.00019-7
Schnotz, W., and Kürschner, C. (2007). A reconsideration of cognitive load theory.
Educ. Psychol. Rev. 19, 469–508. doi: 10.1007/s10648-007-9053-4
Schwab, J. F., and Somerville, L. H. (2022). Raising the stakes for online
learning: monetary incentives increase performance in a computer-based learning
task under certain conditions. Front. Psychol. 13:301. doi: 10.3389/
fpsyg.2022.780301
Shabani, K., Khatib, M., and Ebadi, S. (2010). Vygotsky's zone of proximal
development: instructional implications and teachers' professional development. Engl.
Lang. Teach. 3, 237–248. doi: 10.5539/elt.v3n4p237
Shemshack, A., and Spector, J. M. (2020). A systematic literature review of
personalized learning terms. Smart Learn. Environ. 7:33. doi: 10.1186/
s40561-020-00140-9
Small, D. M., Gitelman, D., Simmons, K., Bloise, S. M., Parrish, T., and
Mesulam, M.-M. (2005). Monetary incentives enhance processing in brain regions
mediating top-down control of attention. Cereb. Cortex 15, 1855–1865. doi: 10.1093/
cercor/bhi063
Spüler, M., Rosenstiel, W., and Bogdan, M. (2012). Online adaptation of a c-VEP
brain-computer interface (BCI) based on error-related potentials and unsupervised
learning. PLoS One 7:e51077. doi: 10.1371/journal.pone.0051077
Steinert, S., Bublitz, C., Jox, R., and Friedrich, O. (2019). Doing things with thoughts:
brain-computer interfaces and disembodied agency. Philos. Technol. 32, 457–482. doi:
10.1007/s13347-018-0308-4
Stipacek, A., Grabner, R., Neuper, C., Fink, A., and Neubauer, A. (2003). Sensitivity of
human EEG alpha band desynchronization to dierent working memory components
and increasing levels of memory load. Neurosci. Lett. 353, 193–196. doi: 10.1016/j.
neulet.2003.09.044
Sweller, J. (1988). Cognitive load during problem solving: eects on learning. Cogn.
Sci. 12, 257–285. doi: 10.1016/0364-0213(88)90023-7
Sweller, J. (1994). Cognitive load theory, learning diculty, and instructional design.
Learn. Instr. 4, 295–312. doi: 10.1016/0959-4752(94)90003-5
Sweller, J. (2010). Element interactivity and intrinsic, extraneous, and germane
cognitive load. Educ. Psychol. Rev. 22, 123–138. doi: 10.1007/s10648-010-9128-5
Sweller, J. (2020). Cognitive load theory and educational technology. Educ. Technol.
Res. Dev. 68, 1–16. doi: 10.1007/s11423-019-09701-3
Sweller, J., van Merrienboer, J. J. G., and Paas, F. G. W. C. (1998). Cognitive architecture
and instructional design. Educ. Psychol. Rev. 10, 251–296. doi: 10.1023/A:1022193728205
Sweller, J., van Merriënboer, J. J. G., and Paas, F. (2019). Cognitive architecture and
instructional design: 20 years later. Educ. Psychol. Rev. 31, 261–292. doi: 10.1007/
s10648-019-09465-5
Tadson, B., Boasen, J., Courtemanche, F., Beauchemin, N., Karran, A.-J., Léger, P.-M.,
et al. (2023). “Neuro-adaptive Interface system to evaluate product recommendations
in the context of E-commerce” in International conference on design science research
in information systems and technology. eds. A. Gerber and R. Baskerville (Cham:
Springer Nature Switzerland), 50–68.
Tang, J., Liu, Y., Hu, D., and Zhou, Z. (2018). Towards BCI-actuated smart wheelchair
system. Bio Med. Eng. Online 17:111. doi: 10.1186/s12938-018-0545-x
Tekin, C., Braun, J., and Schaar, M. V. D. (2015). eTutor: online learning for
personalized education. 2015 IEEE international conference on acoustics, speech and
signal processing (ICASSP), (pp.5545–5549). IEEE.
Tetzla, L., Schmiedek, F., and Brod, G. (2021). Developing personalized education:
a dynamic framework. Educ. Psychol. Rev. 33, 863–882. doi: 10.1007/s10648-020-09570-w
Tohidi, H., and Jabbari, M. M. (2012). e eects of motivation in education. Procedia
Soc. Behav. Sci. 31, 820–824. doi: 10.1016/j.sbspro.2011.12.148
Torres-García, A. A., Martínez-Santiago, F., Montejo-Ráez, A., and Ureña-López, L. A.
(2023). Toward an educative EEG-based neuroIIR system for adapting contents. Int. J.
Hum. Comput. Interact., 1–15. doi: 10.1080/10447318.2023.2275088
Tuladhar, A. M., Huurne, N., Schoffelen, J. M., Maris, E., Oostenveld, R., and
Jensen, O. (2007). Parieto-occipital sources account for the increase in alpha
activity with working memory load. Hum. Brain Mapp. 28, 785–792. doi: 10.1002/
hbm.20306
Urigüen, J. A., and Garcia-Zapirain, B. (2015). EEG artifact removal—state-of-the-art
and guidelines. J. Neural Eng. 12:031001. doi: 10.1088/1741-2560/12/3/031001
Van der Kleij, F. M., Feskens, R. C., and Eggen, T. J. (2015). Eects of feedback in a
computer-based learning environment on students’ learning outcomes: a meta-analysis.
Rev. Educ. Res. 85, 475–511. doi: 10.3102/0034654314564881
van Merriënboer, J. J. G., and Ayres, P. (2005). Research on cognitive load theory and
its design implications for e-learning. Educ. Technol. Res. Dev. 53, 5–13. doi: 10.1007/
BF02504793
Vansteensel, M. J., Klein, E., van iel, G., Gaytant, M., Simmons, Z., Wolpaw, J. R.,
et al. (2023). Towards clinical application of implantable brain–computer interfaces for
people with late-stage ALS: medical and ethical considerations. J. Neurol. 270,
1323–1336. doi: 10.1007/s00415-022-11464-6
Värbu, K., Muhammad, N., and Muhammad, Y. (2022). Past, present, and future of
EEG-based BCI applications. Sensors 22:3331. doi: 10.3390/s22093331
Venthur, B., Blankertz, B., Gugler, M. F., and Curio, G. (2010). Novel applications of
BCI technology: psychophysiological optimization of working conditions in industry.
2010 IEEE international conference on systems, man and cybernetics,
(pp.417–421). IEEE.
Verkijika, S. F., and De Wet, L. (2015). Using a brain-computer interface (BCI) in
reducing math anxiety: evidence from SouthAfrica. Comput. Educ. 81, 113–122. doi:
10.1016/j.compedu.2014.10.002
Vijayalakshmi, R., Nandagopal, D., Dasari, N., Cocks, B., Dahal, N., and ilaga, M.
(2015). Minimum connected component–a novel approach to detection of cognitive
load induced changes in functional brain networks. Neurocomputing 170, 15–31. doi:
10.1016/j.neucom.2015.03.092
Vincent, J. L., Kahn, I., Snyder, A. Z., Raichle, M. E., and Buckner, R. L. (2008).
Evidence for a frontoparietal control system revealed by intrinsic functional connectivity.
J. Neurophysiol. 100, 3328–3342. doi: 10.1152/jn.90355.2008
Vlachogianni, P., and Tselios, N. (2022). Perceived usability evaluation of educational
technology using the system usability scale (SUS): a systematic review. J. Res. Technol.
Educ. 54, 392–409. doi: 10.1080/15391523.2020.1867938
Vygotsky, L. S., and Cole, M. (1978). Mind in society: development of higher
psychological processes. Cambridge, Massachusetts: Harvard University Press.
Walter, C., Rosenstiel, W., Bogdan, M., Gerjets, P., and Spüler, M. (2017). Online EEG-
based workload adaptation of an arithmetic learning environment. Front. Hum.
Neurosci. 11:286. doi: 10.3389/fnhum.2017.00286
Wang, S., Gwizdka, J., and Chaovalitwongse, W. A. (2016). Using wireless EEG signals
to assess memory workload in the n-Back task. IEEE Transact. Hum. Mach. Syst. 46,
424–435. doi: 10.1109/THMS.2015.2476818
Wang, H., and Lehman, J. D. (2021). Using achievement goal-based personalized
motivational feedback to enhance online learning. Educ. Technol. Res. Dev. 69, 553–581.
doi: 10.1007/s11423-021-09940-3
Wang, Y., Nakanishi, M., and Zhang, D. (2019). “EEG-based brain-computer
interfaces” in Neural Interface: Frontiers and applications. ed. X. Zheng (Singapore:
Springer), 41–65.
Wickersham, L. E., and McGee, P. (2008). Perceptions of satisfaction and deeper
learning in an online course. Quart. Rev. Dist. Educ. 9:73.
Wimmer, G. E., and Poldrack, R. A. (2022). Reward learning and working memory:
eects of massed versus spaced training and post-learning delay period. Mem. Cogn. 50,
312–324. doi: 10.3758/s13421-021-01233-7

Beauchemin et al. 10.3389/fnhum.2024.1416683
Frontiers in Human Neuroscience 23 frontiersin.org
Wouters, P., Paas, F., and van Merriënboer, J. J. (2008). How to optimize learning from
animated models: a review of guidelines based on cognitive load. Rev. Educ. Res. 78,
645–675. doi: 10.3102/0034654308320320
Wu, W., Wang, B., Zheng, W., Liu, Y., and Yin, L. (2020). Higher education online
courses personalized recommendation algorithm based on score and attributes. J. Phys.
Conf. Ser. 1673:12025. doi: 10.1088/1742-6596/1673/1/012025
Xia, Q., Chiu, T. K., and Li, X. (2023). A scoping review of BCIs for learning regulation
in mainstream educational contexts. Behav. Inform. Technol. 43, 1–22. doi:
10.1080/0144929X.2023.2241559
Xiao, J., Wang, M., Jiang, B., and Li, J. (2018). A personalized recommendation system
with combinational algorithm for online learning. J. Ambient. Intell. Humaniz. Comput.
9, 667–677. doi: 10.1007/s12652-017-0466-8
Xu, K. M., Koorn, P., de Koning, B., Skuballa, I. T., Lin, L., Henderikx, M., et al.
(2021). A growth mindset lowers perceived cognitive load and improves learning:
integrating motivation to cognitive load. J. Educ. Psychol. 113, 1177–1191. doi: 10.1037/
edu0000631
Yuksel, B. F., Oleson, K. B., Harrison, L., Peck, E. M., Afergan, D., Chang, R., et al.
(2016). Learn piano with BACh: An adaptive learning interface that adjusts task
diculty based on brain state. Proceedings of the 2016 CHI conference on human
factors in computing systems, (pp.5372–5384).
Zammouri, A., Moussa, A. A., and Mebrouk, Y. (2018). Brain-computer interface for
workload estimation: assessment of mental eorts in learning processes. Expert Syst.
Appl. 112, 138–147. doi: 10.1016/j.eswa.2018.06.027
Zander, T. O., and Kothe, C. (2011). Towards passive brain–computer interfaces:
applying brain–computer interface technology to human–machine systems in general.
J. Neural Eng. 8:025005. doi: 10.1088/1741-2560/8/2/025005
Zander, T. O., Kothe, C., Welke, S., and Roetting, M. (2009). “Utilizing secondary
input from passive brain-computer interfaces for enhancing human-machine
interaction” in Foundations of Augmented Cognition eds. D. D. Schmorrow, I. V.
Estabrooke, and M. Grootjen (Berlin, Heidelberg: Neuroergonomics and Operational
Neuroscience).
Zhou, Y., Xu, T., Cai, Y., Wu, X., and Dong, B. (2017a). Monitoring cognitive
workload in online videos learning through an EEG-based brain-computer
Interface. Cham: Learning and Collaboration Technologies. Novel Learning
Ecosystems.
Zhou, Y., Xu, T., Cai, Y., Wu, X., and Dong, B. (2017b). Monitoring cognitive workload
in online videos learning through an EEG-based brain-computer interface. Learning
and collaboration technologies. Novel learning ecosystems: 4th international conference,
LCT 2017, held as part of HCI international 2017, Vancouver, BC, Canada, July 9–14,
2017, proceedings, part I4, (pp.64–73). Springer International Publishing.