Enhancing Learning with Physiological Measures: A Systematic

Review of Applications in Neuroeducation

Blaha Gregory Correia dos Santos Goussain

1,2 a

, Roque Antônio de Moura

José Roberto Dale Luche

, Herlandí de Souza Andrade

and Messias Borges Silva

1,3 e

São Paulo State University (UNESP), School of Engineering and Sciences, Guaratinguetá, São Paulo, Brazil

University of Tennessee, Department of Industrial and Systems Engineering, Knoxville, TN, U.S.A.

University of São Paulo (USP), School of Engineering of Lorena, Lorena, São Paulo, Brazil

Keywords: Neuroeducation, Student Engagement, Electrodermal Activity, Physiological Measures, Educational

Neuroscience.

Abstract: This systematic review explores the integration of neuroscience and education, focusing on physiological

monitoring technologies such as Electrodermal Activity (EDA), Heart Rate (HR), and Skin Temperature (ST).

These metrics, facilitated by wearable devices and machine learning models, provide real-time insights into

student engagement, emotional states, and academic performance. The analysis synthesizes findings from

recent studies, highlighting the transformative potential of physiological measures in creating adaptive,

student-centered learning environments. The review examines the use of physiological monitoring in

education for stress assessment, motivation enhancement, and academic performance optimization, while also

addressing challenges in reliability, ethics, and implementation. By identifying existing gaps, it proposes

directions for future research to refine these tools and promote their widespread adoption in educational

contexts. These advancements underscore the role of physiological insights in fostering emotional well-being

and optimizing teaching practices, marking a significant step toward evidence-based, neuroeducation-

informed strategies.

1 INTRODUCTION

The intersection of neuroscience and education offers

a promising avenue for optimizing learning

environments. Understanding physiological

processes allows educators to tailor teaching methods

to students’ cognitive and emotional needs. Key

metrics such as Electrodermal Activity (EDA), Heart

Rate (HR), and Skin Temperature (ST) have emerged

as critical tools for real-time insights into student

engagement, stress, and emotional states. Leveraging

these metrics, studies have highlighted the potential

of physiological monitoring in enhancing teaching

and learning practices. These dynamic measures hold

the potential to enhance learning outcomes through a

deeper understanding of student behavior and

https://orcid.org/0000-0002-6325-9410

https://orcid.org/0000-0002-3036-7116

https://orcid.org/0000-0001-5302-7301

https://orcid.org/0000-0003-3293-3991

https://orcid.org/0000-0002-8656-0791

performance (Amaral & Fregni, 2021). Furthermore,

Moura et al. (2022) highlight the imperative for

educational and corporate strategies to address skill

gaps and workforce demands, particularly within the

context of Industry 4.0, which necessitates a blend of

technical expertise, creativity, and collaboration.

Despite growing research, a comprehensive

review of physiological monitoring in education is

lacking. This study examines its applications,

benefits, challenges, and reliability in adaptive

learning. A systematic search in major databases

identified peer-reviewed studies on EDA, HR, and ST

as engagement indicators.

Relevant studies applying physiological

monitoring in education were selected, while non-

educational research was excluded. Extracted data

Goussain, B. G. C. S., Moura, R. A., Luche, J. R. D., Andrade, H. S. and Silva, M. B.

Enhancing Learning with Physiological Measures: A Systematic Review of Applications in Neuroeducation.

DOI: 10.5220/0013438400003932

In Proceedings of the 17th International Conference on Computer Supported Education (CSEDU 2025) - Volume 1, pages 111-122

ISBN: 978-989-758-746-7; ISSN: 2184-5026

111

included study design, sample characteristics,

physiological measures, and key findings. A

qualitative synthesis identified trends and gaps, with

a methodological appraisal assessing data reliability.

This review critically evaluates physiological

monitoring in education, highlighting its potential,

limitations, and integration strategies.

1.1 Physiological Metrics in Education

EDA has been extensively studied as a reliable

indicator of academic performance and engagement.

For instance, Horvers et al. (2021) highlighted its

efficacy in monitoring and predicting students’

participation in academic tasks. Building on this,

Abromavičius et al. (2023) demonstrated the utility of

combining EDA, HR, and ST metrics to predict

academic performance, emphasizing the importance

of feature selection and advanced modeling

techniques in educational research.

The application of neurophysiological metrics

extends beyond engagement to address motivation

and cognitive retention. Sánchez-Carracedo et al.

(2021) demonstrated that neuroscience-based

strategies can enhance students’ motivation and

attention, leading to improved conceptual retention.

Similarly, Khan et al. (2019) examined the

correlations between physiological responses, such as

EDA and ST, and their impact during high-pressure

academic tasks, providing evidence of their relevance

in challenging learning environments.

1.2 Multimodal and Emotional

Engagement Approaches

Research has increasingly explored the value of

multimodal approaches in active learning scenarios.

Villanueva et al. (2018) identified the potential of

combining EDA with other modalities to enhance

engagement and academic performance in active

learning contexts. Meanwhile, Loderer et al. (2020)

established a link between emotional engagement and

improved outcomes in technology-based learning

environments. These findings align with the work of

Thammasan et al. (2020), who demonstrated the

feasibility of monitoring physiological signals such as

EDA and HR through wearable sensors, making these

insights more accessible in real-time educational

settings.

Integrating emotional and physiological metrics

into pedagogical strategies has been shown to enrich

learning experiences. For example, Eliot and Hirumi

(2019) advocated for the inclusion of emotional

engagement measures to enhance educational

practices and foster more personalized learning

environments. Similarly, Darvishi et al. (2022)

highlighted the potential of neurophysiological

measures to address individual cognitive and

emotional needs, thereby improving overall

educational practices.

Collectively, these studies underscore the critical

role of physiological insights in advancing the field

of neuroeducation. By leveraging neurophysiological

data, educators can create adaptive, student-centered

environments that promote both academic success

and emotional well-being. This research highlights

the intersection of neuroscience and education as a

fertile ground for innovation, offering a robust

framework to enhance traditional and technology-

mediated learning.

2 PHYSIOLOGICAL MEASURES

IN EDUCATION

2.1 Electrodermal Activity

EDA, a measure of sympathetic nervous system

activity, offers valuable insights into emotional

arousal and cognitive states by quantifying variations

in skin conductance, typically measured in

microsiemens (μS). This metric has gained

prominence in education research for its ability to

provide objective, real-time data on student

engagement, stress, and emotional responses.

2.1.1 Advancements in EDA Measurement

Techniques

Several studies have contributed to refining EDA

measurement methodologies. Quintero et al. (2016b)

introduced the TVSymp index, utilizing time-

frequency spectral analysis to enhance the

consistency of sympathetic activity assessments.

Similarly, Quintero et al. (2016a) demonstrated the

significance of low-frequency EDA components

(0.045–0.15 Hz) in evaluating responses to cognitive

and physical stressors.

Geršak and Drnovšek (2020) developed a

simulator to improve the precision of metrological

evaluations for EDA devices, advancing the

reliability of EDA measurements. Additionally,

Hernando-Gallego et al. (2017) proposed the

SparsEDA algorithm, which improved computational

efficiency and interpretability of EDA data.

Nourbakhsh et al. (2012) validated the

relationship between EDA and cognitive load,

highlighting spectral features' ability to differentiate

CSEDU 2025 - 17th International Conference on Computer Supported Education

112

task difficulties. These advancements have

significantly enhanced the accuracy and utility of

EDA measurements, making them more practical for

educational and non-educational applications alike.

2.1.2 Applications in Stress and Emotional

Analysis

EDA has been particularly useful in analyzing stress

and emotional responses. Lui and Du (2018)

developed a method for psychological stress

detection based solely on EDA, achieving an 81.82%

recognition rate using Fisher projection and linear

discriminant analysis.

Sánchez-Reolid et al. (2020) combined EDA with

other physiological signals, achieving up to 99.69%

accuracy in stress detection using neural networks

and Adaboost algorithms. Villarejo et al. (2012)

created a wearable stress sensor based on galvanic

skin response (GSR), demonstrating a 76.56%

success rate. Malathi et al. (2018) extended the

application of EDA to road safety, developing a

device for real-time drowsiness detection in drivers.

Poh et al. (2010) validated wearable sensors for

continuous EDA assessment, identifying consistent

patterns of sympathetic modulation during daily

activities. These studies underscore EDA's role in

stress monitoring and its broader applicability in

diverse contexts beyond education.

2.1.3 EDA in Educational Contexts

In educational environments, EDA has demonstrated

significant potential for enhancing teaching and

learning practices. Di Lascio et al. (2018) used

wearable EDA sensors to distinguish engaged

students from disengaged ones, achieving 81% recall

using support vector machines (SVM).

Villanueva et al. (2019) explored EDA responses

in academic mentoring settings, revealing the

influence of identity on physiological responses. Reid

et al. (2020) combined EDA data with behavioral

analyses to identify key factors affecting academic

performance, demonstrating the value of integrating

physiological and qualitative data.

Pijeira-Díaz et al. (2018) employed EDA to detect

moments of high and low engagement in classroom

settings, offering insights into students' emotional and

cognitive states. Villanueva et al. (2018) incorporated

EDA into multimodal assessments during

engineering activities, showing increased EDA levels

during active, collaborative tasks.

Potter et al. (2019) highlighted EDA’s ability to

gauge student engagement across various teaching

methodologies, emphasizing its utility for real-time

feedback. These studies illustrate the versatility of

EDA as a tool for assessing engagement, emotional

states, and the effectiveness of educational

interventions. Its integration into multimodal

approaches has proven especially valuable in active

and collaborative learning scenarios.

2.1.4 Implications for Pedagogical Strategies

The application of EDA in education provides a non-

invasive and objective method for monitoring

students' physiological responses. By bypassing the

biases often associated with self-reported measures

(Caruelle et al., 2019), EDA enables educators to

tailor pedagogical strategies more effectively. From

real-time feedback to long-term performance

monitoring, EDA contributes to a nuanced

understanding of student engagement and learning

outcomes.

The growing body of research on EDA highlights

its relevance in neuroeducation, offering robust

methods for measuring engagement and emotional

states. By integrating these insights into teaching

practices, educators can create adaptive, data-driven

environments that enhance both academic

performance and emotional well-being.

2.2 Heart Rate

HR serve as indicators of physiological arousal and

stress, providing valuable insights into students'

engagement and emotional states. Schneider et al.

(2020) explored HR synchronization among

collaborators during programming tasks, revealing

positive correlations between synchronization and

task performance. This study highlights the potential

of HR metrics as objective measures of collaboration

quality and engagement in group activities.

Similarly, Ghannam et al. (2020) emphasized the

role of HR monitoring in neuroengineering education,

demonstrating how wearable technologies can

enhance learning experiences by offering real-time

physiological feedback.

2.2.1 Applications in Active Learning and

Physical Engagement

Research has also focused on the role of HR in active

learning contexts. Darnell and Krieg (2019) analyzed

HR fluctuations during active learning sessions,

identifying strong correlations between physiological

engagement and academic interaction. Their findings

underscore the importance of HR monitoring as a tool

for assessing and optimizing engagement in dynamic

educational settings.

Enhancing Learning with Physiological Measures: A Systematic Review of Applications in Neuroeducation

113

Wang and Liu (2019) extended this research to

physical education, utilizing wearable devices to

monitor HR and provide real-time feedback. Their

findings revealed significant differences in

engagement levels among participants, showcasing

the potential of wearable technology in fostering

personalized education strategies and enhancing

student participation.

2.2.2 Consistency and Contextual Analysis

of HR and HRV

The consistency of HR and and Heart Rate Variability

(HRV) measures has also been a topic of

investigation. Quintero and Bolkhovsky (2019)

examined these metrics under controlled conditions,

identifying low HRV indices as reliable markers of

physiological engagement. Their research highlights

the utility of HRV as a robust measure for evaluating

focus and stress levels in educational settings.

Additionally, Gao et al. (2020) combined HR data

with environmental variables to predict student

engagement in diverse classroom environments.

Their work demonstrates the value of integrating

physiological and contextual data to improve

teaching methods and outcomes.

Collectively, these studies establish HR and HRV

as essential tools for understanding and enhancing

educational outcomes. By integrating these measures

into classroom practices, educators and researchers

can gain a deeper understanding of physiological

engagement, allowing for more targeted and effective

interventions that enhance both academic

performance and emotional well-being.

2.3 Skin Temperature

ST is a subtle but impactful physiological indicator,

reflecting the body’s thermoregulation processes,

which are influenced by emotional and environmental

factors.

Studies such as Pérez et al. (2018) have

demonstrated significant correlations between ST and

academic performance, particularly under stress. This

research highlights the relevance of ST as a non-

invasive marker for understanding students’

emotional and cognitive states during high-pressure

academic activities. However, Terriault et al. (2021)

identified challenges in real-time ST monitoring

during educational activities, particularly due to

external environmental factors that can affect

measurement accuracy.

2.3.1 Wearable Technology and Real-Time

Monitoring

The development of wearable technology has

advanced the continuous monitoring of ST,

facilitating its application in educational settings.

Yoon et al. (2016) emphasized the utility of wearable

sensors, demonstrating their effectiveness for real-

time stress detection and intervention. Their findings

showcase the potential of wearable patches for long-

term ST monitoring, enabling educators to better

understand students' physiological responses in

diverse learning environments. Additionally, Pérez et

al. (2018) explored the application of wearable

devices in assessing stress levels among students,

finding significant correlations between ST and

academic performance during high-stress tasks.

2.3.2 Multimodal Approaches Combining

Skin Temperature

Combining ST with other physiological measures has

further enhanced its predictive power in

understanding emotional and cognitive states.

Rodríguez-Arce et al. (2020) demonstrated that

integrating ST with metrics such as HR and EDA can

accurately predict stress and anxiety levels in

academic settings. This multimodal approach

provides a more comprehensive understanding of

how physiological signals interact, offering insights

into student behavior and well-being. These findings

underscore the critical role of ST as an indicator of

emotional and physiological states, with practical

implications for creating adaptive and supportive

learning environments. By leveraging advances in

wearable technology and combining ST with other

physiological measures, educators can develop

strategies that address students' emotional needs,

enhance academic performance, and foster a more

inclusive and responsive educational experience.

3 METHODOLOGICAL

ADVANCEMENTS

3.1 Wearable Technology

The integration of wearable devices, such as the

Empatica E4, has revolutionized physiological

monitoring in educational contexts by enabling

unobtrusive and real-time data collection. These

devices capture multimodal metrics, including EDA,

HR, and ST, making them highly applicable for

classroom environments.

CSEDU 2025 - 17th International Conference on Computer Supported Education

114

3.1.1 Physiological Monitoring and Data

Collection

Research by Rodic-Trmcic et al. (2016) emphasized

the role of wearable solutions in assessing

physiological arousal and engagement within

classroom settings. Their findings highlighted the

utility of Skin Conductance Response (SCR) and HR

data for monitoring stress and underscored the

importance of mobile assessment systems in

delivering continuous feedback to improve

educational quality. Building on this, Lu et al. (2017)

proposed a framework leveraging widely available

wearable devices to monitor basic student actions and

infer engagement levels through sensors such as

accelerometers and HR monitors. This framework

demonstrated how wearable technology can capture

both physiological and behavioral data to enhance the

understanding of student dynamics.

The study by Domínguez-Jiménez et al. (2020)

advanced the application of wearable devices by

integrating GSR and Photoplethysmogram (PPG)

signals for emotion recognition, achieving a precision

rate of up to 100%. Pérez et al. (2018) also

highlighted the practical utility of wearable sensors

for real-time stress monitoring among students,

demonstrating their capacity to support adaptive

educational strategies. Similarly, Yoon et al. (2016)

contributed to the development of flexible wearable

patches capable of continuously monitoring EDA,

HR, and ST, providing a robust solution for long-term

use in educational settings.

3.1.2 Adaptive Interventions and

Personalized Education

Rodríguez-Arce et al. (2020) demonstrated the

reliability of wearable devices in accurately predicting

stress and anxiety levels, particularly in high-pressure

academic environments. Their findings underscored

the robustness of wearable technology in measuring

physiological responses critical for student well-being.

Gao et al. (2020) extended this research by integrating

wearable devices with environmental data to enhance

engagement predictions. This approach not only

provided more comprehensive insights but also offered

actionable data to educators for tailoring interventions

to individual student needs. These methodological

advancements illustrate the transformative potential of

wearable technology in modern education. By enabling

continuous, precise, and real-time monitoring of

physiological and contextual data, these devices are

paving the way for a more adaptive, personalized, and

effective educational experience.

3.2 Machine Learning Integration

Machine learning (ML) techniques have

revolutionized the analysis of physiological data,

enabling the development of predictive models for

student engagement and academic performance. For

example, Pérez et al. (2018) demonstrated how

combining ML algorithms with multimodal

physiological data could effectively detect stress,

highlighting the potential of these techniques in

educational settings. Similarly, Cain and Lee (2016)

applied ML methods to Makerspace activities,

identifying moments of high engagement by

correlating them with peaks in EDA and HR data.

Kanna et al. (2018) showcased a practical integration

of wearable sensors in engineering education,

allowing students to analyze their own physiological

data, such as ECG signals, while learning signal

processing techniques.

3.2.1 Machine Learning for Student

Performance Prediction

Supervised learning techniques have proven

particularly effective in predicting student

performance. Rastrollo-Guerrero et al. (2020)

demonstrated the utility of Support Vector Machines

(SVM), achieving high accuracy in performance

prediction. Simjanoska et al. (2014) expanded this

application by using ML algorithms to develop

adaptive e-Learning strategies, ensuring targeted

learning outcomes while reducing random guessing.

Ensemble methods, such as RealAdaBoost combined

with J48, were shown by Imran et al. (2019) to

improve model precision, achieving a classification

accuracy of 95.78%. Walsh and Mahesh (2017)

integrated behavioral and traditional academic data to

predict outcomes early, enabling timely interventions

that significantly improved learning experiences.

3.2.2 Leveraging Diverse Machine Learning

Techniques

Various ML algorithms have been applied to enhance

prediction models across educational settings. Pavani

et al. (2017) highlighted Decision Tree (DT)

algorithms, such as C4.5, for their accessibility and

effectiveness in predicting academic performance.

Shanthini et al. (2018) demonstrated the potential

of ensemble methods, including AdaBoost and

Bagging, achieving accuracy rates of 97.6%. Yan and

Liu (2020) validated the superiority of stacking

models, which combine algorithms like Random

Enhancing Learning with Physiological Measures: A Systematic Review of Applications in Neuroeducation

115

Forests (RF), SVM, and AdaBoost, to improve

predictive accuracy

Ofori et al. (2020) incorporated socio-economic

factors into their ML models, revealing significant

impacts on academic outcomes, while Naicker et al.

(2020) compared Linear SVM (LSVM) with other

algorithms to identify its effectiveness across diverse

student demographics.

3.2.3 Early Identification and Adaptive

Learning Systems

ML has also been employed for early identification of

at-risk students. Wakelam et al. (2019) applied RF

and K-Nearest Neighbors (KNN) algorithms in small-

class settings to predict academic challenges,

achieving high reliability. Hussain et al. (2018)

demonstrated ML integration in real-time learning

systems, enabling continuous feedback and adaptive

interventions. Polyzou and Karypis (2023)

emphasized the use of Gradient Boosting and RF

models for early warning systems, while Pang et al.

(2017) achieved high accuracy in graduation

predictions by incorporating psychopedagogical

variables into SVM-based models.

Gray and Perkins (2019) successfully identified

at-risk students by the third week of a semester using

ML models with a 97% accuracy rate, and Zabriskie

et al. (2019) employed RF models to develop early-

warning systems in physics courses, leveraging

institutional and classroom data.

3.2.4 Machine Learning in Physiological

Data Analysis

Integrating ML with physiological data has opened

new possibilities for adaptive education. Gao et al.

(2020) combined ML techniques with wearable

technology to analyze multimodal physiological data,

improving engagement predictions in diverse

learning environments.

Yoon et al. (2016) highlighted the feasibility of

ML models for continuous data stream analysis,

allowing educators to tailor strategies based on real-

time feedback. Pérez et al. (2018) emphasized the

effectiveness of ML in optimizing stress detection,

enabling adaptive interventions to support student

well-being. These advancements underscore the role

of ML in leveraging physiological insights to create

personalized and effective educational experiences,

making adaptive learning environments more feasible

and impactful.

4 IMPLICATIONS FOR

NEUROEDUCATION

The integration of physiological measures into

neuroeducation has opened new possibilities for

understanding and enhancing individual learning

processes. By addressing both cognitive and

emotional dimensions, educators can create inclusive

and adaptive learning environments that cater to

diverse student needs.

Abromavičius et al. (2023) emphasized the

practical implications of using physiological data to

manage stress and enhance academic outcomes,

particularly in high-pressure contexts. Similarly,

Schneider et al. (2020) explored physiological

synchronization metrics as indicators of effective

teamwork in collaborative learning environments,

highlighting their potential to improve group

dynamics and performance.

Table 1 provides a comprehensive summary of

empirical findings from 17 key studies that

investigated the application of physiological

measures, including EDA, HR, and ST, in educational

contexts. These studies span diverse experimental

settings, from traditional classrooms to e-learning

environments, and highlight the potential of these

metrics for monitoring engagement, stress, and

emotional states. The table synthesizes data on

participants, experimental conditions, and significant

outcomes, offering valuable insights into the practical

applications and limitations of physiological

monitoring technologies.

The studies summarized in Table 1 underscore the

versatility and efficacy of physiological measures in

enhancing educational practices. Key findings reveal

that EDA consistently emerges as a reliable indicator

of student engagement, as demonstrated by Di Lascio

et al. (2018) and Villanueva et al. (2019), with

engagement detection accuracies reaching up to 81%

using advanced machine learning models. Similarly,

HR and ST have been validated as complementary

measures, particularly in stress-inducing academic

environments, with studies such as Pérez et al. (2018)

showcasing their predictive value in estimating stress

levels with high precision.

One notable trend across the studies is the

increasing reliance on multimodal approaches that

integrate EDA, HR, and ST to achieve more robust

insights. For instance, Rodríguez-Arce et al. (2020)

demonstrated a stress detection accuracy of 90% by

combining these metrics, highlighting the synergistic

potential of multimodal data in understanding complex

physiological responses during academic tasks.

CSEDU 2025 - 17th International Conference on Computer Supported Education

116

Table 1: Studies on EDA, HR, and ST in education: participants, conditions, and key findings.

References Title Partici

ants Stresso

Results

Abromavičiu

s et al. (2023)

Prediction of exam scores

using a multi-sensor

approach for wearable

exam stress dataset with

uniform preprocessing

undergraduate

students

Exam stress

during three

examinations

Physiological signals, including EDA,

HR, and ST, revealed high predictive

potential for exam scores with accuracy,

AUROC, and F1-score reaching 0.9,

0.89, and 0.87, respectively. A uniform

preprocessing enhanced the robustness of

nal anal

sis.

Al-Awani

(2016)

A Combined Approach to

Improve Supervised E-

Learning using Multi-

Sensor Student

Engagement Analysis

20 students

E-learning

sessions

Correlation analysis of EDA, pulse rate,

and facial expressions indicated

significant potential for measuring

engagement. Findings suggest integration

of multi-sensor data to dynamically

ust educational content.

Cain & Lee

(2016)

Measuring electrodermal

activity to capture

engagement in an

afterschool maker progra

2 youth

participants

Practical

activities in a

makerspace

Analysis of EDA data indicated higher

engagement during interactive activities,

such as presenting progress, and varied

engagement during individual tasks.

Darnell &

Krieg (2019)

Student Engagement

Assessed Using Heart Rate

Shows No Reset

Following Active

Learning Sessions in

Lectures

15 students

Lecture-based

active

learning

sessions

HR increased during active learning but

returned to baseline immediately

afterward. Demonstrated HR's limitations

in reflecting sustained engagement post-

activity.

Di Lascio et

al. (2018)

Unobtrusive Assessment

of Students’ Emotional

Engagement during

Lectures Using

Electrodermal Activity

Sensors

24 students

and 9

professors in

41 lectures

Emotional

engagement

during

lectures

EDA sensors identified disengaged

students with 81% accuracy using SVM,

highlighting the potential of EDA for

educational feedback.

Gao et al.

(2020)

n-Gage: Predicting In-

Class Emotional,

Behavioral, and Cognitive

Engagement in the Wild

23 students

(13 females

and 10 males)

and 6 teachers

(4 females and

2 males)

Classroom

engagement

tasks

Multidimensional engagement prediction

(emotional, behavioral, and cognitive)

using EDA, HRV, and ST achieved MAE

of 0.788 and RMSE of 0.975.

Highlighted the utility of wearable

sensors for real-time engagement

monitoring.

Jamal &

Kamioka

(2019)

Emotions Detection

Scheme Using Facial Skin

Temperature and Heart

Rate Variability

20 subjects

(10 females

and 10 males)

Visual and

auditory

stimuli

Emotion detection (joy, fear, sadness, and

relaxation) using HRV and facial skin

temperature achieved 88.75% accuracy

with an ANN-based classifier.

Demonstrated the reliability of HRV and

facial skin temperature for emotion

recognition without physical interaction.

Khan et al.

(2019)

Exploring relationships

between electrodermal

activity, skin temperature,

and performance during

engineering exams

76 engineering

students

Exam

difficulty and

cognitive

tasks

Weak but significant correlations

observed between EDA, ST, and exam

difficulty index (e.g., r=0.13 for EDA;

r=0.08 for ST). Regression models

indicated moderate significance in

relationships among variables.

Lu et al.

(2017)

A Framework for Learning

Analytics Using

Commodity Wearable

Devices

24 participants

(11 female and

13 male)

Academic

stress and

physical

activity

Developed the LEARNSense framework

integrating EDA, HR, and ST data for

analyzing student engagement. Achieved

F1 scores of 0.9 for classifying

engagement states. Demonstrated

feasibility of wearable sensors for real-

time analytics.

Enhancing Learning with Physiological Measures: A Systematic Review of Applications in Neuroeducation

117

Table 1: Studies on EDA, HR, and ST in education: participants, conditions, and key findings (cont.).

References Title Partici

ants Stresso

Results

Nourbakhsh

et al. (2012)

Using galvanic skin

response for cognitive load

measurement in arithmetic

and reading tasks

25 participants

(13 for

arithmetic

tasks, 12 for

readin

tasks

)

Reading and

arithmetic

tasks of

varying

difficult

Spectral features of GSR demonstrated

high significance in cognitive load

measurement after normalization,

highlighting the potential of spectral

anal

sis for com

lex co

nitive tasks.

Pérez et al.

(2018)

Evaluation of

Commercial-Off-The-

Shelf Wrist Wearables to

Estimate Stress on

Students

12 first-year

university

students

Stress-

inducing

laboratory

tasks and

classroom

activities

Protocol validated the efficacy of COTS

wearables in capturing HR, HRV, and ST

for stress analysis in educational settings.

Machine learning models demonstrated

high accuracy in estimating stress levels

durin

academic tasks.

Pijeira-Díaz

et al. (2018)

Profiling sympathetic

arousal in a physics course

how active are students

24 high school

students

Advanced

physics

course and

final exam

Arousal measured via EDA positively

correlated with academic performance (r

= 0.66). Low arousal states were

predominant during lectures, while

activation significantly increased during

the exam.

Quintero &

Bolkhovsky

(2019)

Machine learning models

for the identification of

cognitive tasks using

autonomic reactions from

heart rate variability and

electrodermal activit

16 participants

(8 male, 8

female)

Cognitive

tasks

including

vigilance and

memory

EDA and HRV indices enabled

identification of cognitive tasks with

classification accuracy up to 66% using

machine learning models like KNN and

SVM.

Rodríguez-

Arce et al.

(2020)

Towards an Anxiety and

Stress Recognition System

for Academic

Environments

21 university

students

Academic

stress tasks

and self-

reported

anxiet

Stress detection achieved 90% accuracy

using k-NN on HR, skin temperature, and

oximetry signals. Anxiety recognition

attained 95% accuracy with SVM using

GSR data.

Schneider et

al. (2020)

Unpacking the relationship

between existing and new

measures of physiological

synchrony and

collaborative learning: a

mixed methods study

42 pairs of

participants

(84

individuals)

Collaborative

tasks

involving

programming

robots

Physiological synchrony (measured via

EDA) correlated with learning gains (r =

0.35) and collaboration quality (r = 0.3).

Developed a novel measure using EDA

cycles that improved correlations with

collaboration quality (r = 0.57).

Terriault et

al. (2021)

Use of electrodermal

wristbands to measure

students' cognitive

engagement in the

classroo

8 participants

(7 students

and 1

professor)

Classes,

workshops,

and exams

EDA data from Empatica E4 wristbands

revealed engagement patterns, but

external factors, such as physical activity

and room temperature, complicated data

consistency.

Thammasan

et al. (2020)

A Usability Study of

Physiological

Measurement in School

Using Wearable Sensors

86 adolescents

in schools

Daily

academic

activities

Demonstrated feasibility of EDA and HR

measurements in school settings.

Addressed challenges in data quality and

preprocessing, highlighting limitations of

eneric si

nal

rocessin

tools.

Additionally, machine learning applications, as seen

in studies like Gao et al. (2020), have further

enhanced the predictive power of these measures,

enabling real-time engagement and emotional

monitoring with high accuracy. Despite these

promising outcomes, the findings also highlight

significant challenges. External factors, such as

environmental conditions and physical activity, can

affect the reliability of HR and ST measurements, as

noted by Terriault et al. (2021). Similarly, the limited

sample sizes in certain studies, such as Cain and Lee

(2016), restrict the generalizability of results,

emphasizing the need for larger-scale investigations.

Moreover, ethical considerations surrounding the use

of wearable devices in educational settings require

further exploration to ensure student privacy and data

security.

In conclusion, the evidence presented in Table 1

underscores the transformative potential of

physiological measures for creating adaptive and

CSEDU 2025 - 17th International Conference on Computer Supported Education

118

student-centered educational environments. By

addressing current limitations and leveraging

advancements in wearable technologies and machine

learning, future research can pave the way for more

inclusive and effective educational practices.

4.1 Practical Applications of

Physiological Insights in Education

The use of physiological data has also been applied to

tailor pedagogical approaches. Villanueva et al.

(2019) demonstrated how the intersection of identity

and physiological metrics in academic mentoring can

address diverse student needs, fostering a more

inclusive learning experience.

Lee et al. (2020) exemplified the use of EDA to

differentiate between public speaking and foreign

language anxiety, showcasing its utility in addressing

distinct stress contexts. Katmada et al. (2015)

proposed a biofeedback system that integrates EDA,

HR, and ST, highlighting its effectiveness in reducing

anxiety through gamified educational tools.

Furthermore, Jamal and Kamioka (2019) introduced

an emotion detection framework using facial ST and

HRV, achieving high accuracy without requiring

physical interaction, thereby expanding the scope of

non-invasive monitoring methods.

4.2 Building Emotionally Supportive

and Adaptive Learning

Environments

Physiological monitoring has proven instrumental in

promoting emotional engagement, which is critical

for deeper learning experiences. Loderer et al. (2020)

demonstrated that emotional engagement, as

measured through physiological data, fosters stronger

connections to educational material and promotes

long-term retention. Reid et al. (2020) highlighted the

value of real-time feedback in identifying stress

points during learning activities, enabling timely and

effective interventions by educators to alleviate stress

and maintain focus. The combination of physiological

metrics has shown promise in creating supportive

frameworks that improve resilience and academic

outcomes, especially in high-stress environments.

Rodríguez-Arce et al. (2020) explored the integration

of EDA, HR, and ST in designing frameworks that

support students' emotional well-being while

fostering academic success.

Eliot and Hirumi (2019) emphasized that

incorporating emotional and physiological metrics

into pedagogy enhances inclusivity and

responsiveness to individual learner needs, paving the

way for more equitable educational practices. These

advancements collectively underscore the

transformative potential of physiological monitoring

in advancing neuroeducation. By leveraging insights

from metrics such as EDA, HR, and ST, educators

can develop data-driven strategies to personalize

learning and address students' emotional and

cognitive challenges. The application of biofeedback

systems, emotion detection frameworks, and real-

time stress monitoring tools highlights the profound

impact of integrating physiological insights into

modern educational practices, ultimately promoting

student success and well-being.

5 CHALLENGES AND FUTURE

DIRECTIONS

The integration of physiological measures into

education presents significant opportunities, yet it

also faces notable challenges. Technical reliability

remains a key concern, particularly for wearable

devices measuring EDA, HR, and ST. External

variables, such as environmental conditions and

physical activity, can affect data accuracy, as

highlighted by studies that emphasize the need for

robust preprocessing techniques and adaptive

algorithms. Moreover, variability in sensor quality

and calibration across devices further complicates

widespread adoption.

The collection of physiological data in

educational settings, especially with minors, raises

significant ethical and privacy concerns. Effective

data governance and security are critical. Future

research must standardize methodologies, enhance

sensor accuracy, and refine machine learning models

to reduce bias in diverse environments. Large-scale,

longitudinal studies are needed for broader validation.

Overcoming these challenges is key to achieving

sustainable educational innovations.

6 CONCLUSIONS

This review underscores the transformative potential

of physiological measures in advancing

neuroeducation. By leveraging metrics such as EDA,

HR, and ST, educators can gain real-time insights into

student engagement, stress, and emotional states,

enabling adaptive and personalized learning

experiences. The integration of wearable

technologies and machine learning models enhances

Enhancing Learning with Physiological Measures: A Systematic Review of Applications in Neuroeducation

119

the feasibility of implementing these approaches in

diverse educational contexts.

Despite the promising applications, challenges

related to technical reliability, ethical considerations,

and scalability remain significant. However, ongoing

advancements in wearable technologies, multimodal

data analysis, and standardized frameworks offer

pathways to address these barriers. Future research

must prioritize inclusivity, ethical transparency, and

empirical validation to ensure the widespread

adoption of these tools.

In conclusion, physiological monitoring

represents a critical innovation in fostering

emotionally supportive, data-driven, and effective

educational environments. By bridging neuroscience

and pedagogy, this approach has the potential to

revolutionize how educators understand and respond

to the cognitive and emotional needs of learners.

ACKNOWLEDGEMENTS

We would like to acknowledge the support of CAPES

for providing the necessary resources and facilities to

conduct this study.

REFERENCES

Abromavicius, V., Serackis, A., Katkevicius, A., Kazlauskas,

M., & Sledevič, T. (2023). Prediction of exam scores

using a multi-sensor approach for wearable exam stress

dataset with uniform preprocessing. Technology and

Health Care, 31(6), 2499–2511. https://doi.org/10.3233/

THC-235015

Al-Alwani, A. (2016). A combined approach to improve

supervised E-learning using multi-sensor student

engagement analysis. American Journal of Applied

Sciences, 13(12), 1377–1384. https://doi.org/10.3844/

ajassp.2016.1377.1384

Amaral, J. A. A., & Fregni, F. (2021). Applying neuroscience

concepts to enhance learning in an online project-based

learning centered course. Journal of Problem-Based

Learning in Higher Education, 9(2), 142–159.

https://doi.org/10.5278/ojs.jpblhe.v9i2.5892

Cain, R., & Lee, V. R. (2016). Measuring electrodermal

activity to capture engagement in an afterschool maker

program. In Proceedings of FabLearn 2016: Conference

on Creativity and Fabrication. ACM.

https://doi.org/10.1145/3003397.3003409

Caruelle, D., Gustafsson, A., Shams, P., & Lervik-Olsen, L.

(2019). The use of electrodermal activity (EDA)

measurement to understand consumer emotions – A

literature review and a call for action. Journal of Business

Research, 104, 146–160. https://doi.org/10.1016/j.jbusr

es.2019.06.041

Darnell, D. K., & Krieg, P. A. (2019). Student engagement,

assessed using heart rate, shows no reset following active

learning sessions in lectures. PLOS ONE, 14(12),

e0225709. https://doi.org/10.1371/journal.pone.0225709

Darvishi, A., Khosravi, H., Sadiq, S., & Weber, B. (2022).

Neurophysiological measurements in higher education:

A systematic literature review. International Journal of

Artificial Intelligence in Education, 32, 413–453.

https://doi.org/10.1007/s40593-021-00256-0

Di Lascio, E., Gashi, S., & Santini, S. (2018). Unobtrusive

assessment of students’ emotional engagement during

lectures using electrodermal activity sensors.

Proceedings of the ACM on Interactive, Mobile,

Wearable and Ubiquitous Technologies, 2(3), 103:1–

103:21. https://doi.org/10.1145/3264913

Domínguez-Jiménez, J. A., Campo-Landines, K. C.,

Martínez-Santos, J. C., Delahoz, E. J., & Contreras-Ortiz,

S. H. (2020). A machine learning model for emotion

recognition from physiological signals. Biomedical

Signal Processing and Control, 55, 101646.

https://doi.org/10.1016/j.bspc.2019.101646

Eliot, J. A. R., & Hirumi, A. (2019). Emotion theory in

education research practice: An interdisciplinary critical

literature review. Educational Technology Research and

Development, 67(5), 1065–1084. https://doi.org/10.10

07/s11423-018-09642-3

Gao, N., Shao, W., Rahaman, M. S., & Salim, F. D. (2020).

n-Gage: Predicting in-class emotional, behavioural and

cognitive engagement in the wild. Proceedings of the

ACM on Interactive, Mobile, Wearable and Ubiquitous

Technologies, 4(3). https://doi.org/10.1145/3411813

Geršak, G., & Drnovšek, J. (2020). Electrodermal activity

patient simulator. PLOS ONE, 15(2), e0228949.

https://doi.org/10.1371/journal.pone.0228949

Ghannam, R., Curia, G., Brante, G., Khosravi, S., & Fan, H.

(2020). Implantable and wearable neuroengineering

education: A review of postgraduate programmes. IEEE

Access, 8, 212395–212407. https://doi.org/10.1109/

ACCESS.2020.3040064

Gray, C. C., & Perkins, D. (2019). Utilizing early

engagement and machine learning to predict student

outcomes. Computers & Education, 131, 22–32.

https://doi.org/10.1016/j.compedu.2018.12.006

Hernando-Gallego, F., Luengo, D., & Artés-Rodríguez, A.

(2017). Feature extraction of galvanic skin responses by

non-negative sparse deconvolution. IEEE Journal of

Biomedical and Health Informatics

. https://doi.org/10.11

09/JBHI.2017.2780252

Horvers, A., Tombeng, N., Bosse, T., Lazonder, A. W., &

Molenaar, I. (2021). Detecting emotions through

electrodermal activity in learning contexts: A systematic

review. Sensors, 21(21), 7869. https://doi.org/10.3390/

s21237869

Hussain, M., Zhu, W., Zhang, W., Abidi, S. M. R., & Ali, S.

(2018). Using machine learning to predict student

difficulties from learning session data. Artificial

Intelligence Review. https://doi.org/10.1007/s10462-

018-9620-8

Imran, M., Latif, S., Mehmood, D., & Shah, M. S. (2019).

Student academic performance prediction using

CSEDU 2025 - 17th International Conference on Computer Supported Education

120

supervised learning techniques. International Journal of

Emerging Technologies in Learning (iJET), 14(14), 4–

12. https://doi.org/10.3991/ijet.v14i14.10310

Jamal, K. M., & Kamioka, E. (2019). Emotions detection

scheme using facial skin temperature and heart rate

variability. MATEC Web of Conferences, 277, 02037.

https://doi.org/10.1051/matecconf/201927702037

Kanna, S., Von Rosenberg, W., Goverdovsky, V.,

Constantinides, A. G., & Mandic, D. P. (2018). Bringing

wearable sensors into the classroom: A participatory

approach. IEEE Signal Processing Magazine, 35(3),

110–116. https://doi.org/10.1109/MSP.2018.2806418

Katmada, A., Chatzakis, M., Apostolidis, H., Mavridis, A., &

Stylianidis, P. (2015). An adaptive serious neuro-game

using a mobile version of a bio-feedback device. In

Proceedings of the International Conference on

Interactive Mobile Communication Technologies and

Learning (IMCL) (pp. 416–420). Thessaloniki, Greece:

IEEE. https://doi.org/10.1109/IMCTL.2015.7359633

Khan, T. H., Villanueva, I., Vicioso, P., & Husman, J. (2019).

Exploring relationships between electrodermal activity,

skin temperature, and performance. Proceedings -

Frontiers in Education Conference, FIE.

https://doi.org/10.1109/FIE43999.2019.9028625

Lee, H., Mandalapu, V., Kleinsmith, A., & Gong, J. (2020).

Distinguishing anxiety subtypes of English language

learners towards augmented emotional clarity. In I. I.

Bittencourt et al. (Eds.), AIED 2020, LNAI 12164 (pp.

157–161). Springer Nature Switzerland AG.

https://doi.org/10.1007/978-3-030-52240-7_29

Lee, V. R., Fischback, L., & Cain, R. (2019). A wearables-

based approach to detect and identify momentary

engagement in afterschool Makerspace programs.

Contemporary Educational Psychology, 59, 101789.

https://doi.org/10.1016/j.cedpsych.2019.101789

Loderer, K., Pekrun, R., & Lester, J. C. (2020). Beyond cold

technology: A systematic review and meta-analysis on

emotions in technology-based learning environments.

Learning and Instruction, 70, 101162.

https://doi.org/10.1016/j.learninstruc.2018.08.002

Lu, Y., Zhang, S., Zhang, Z., Xiao, W., & Yu, S. (2017). A

framework for learning analytics using commodity

wearable devices. Sensors, 17(6), 1382.

https://doi.org/10.3390/s17061382

Lui, Y., & Du, S. (2018). Psychological stress level detection

based on electrodermal activity. Behavioural Brain

Research, 341, 50–53. https://doi.org/10.1016/j.bbr.20

17.12.021

Malathi, D., Dorathi Jayaseeli, J. D., Madhuri, S., &

Senthilkumar, K. (2018). Electrodermal activity-based

wearable device for drowsy drivers. Journal of Physics:

Conference Series, 1000. https://doi.org/10.1088/1742-

6596/1000/1/012048

Moura, R., Richetto, M., Luche, D., Tozi, L., & Silva, M.

(2022). New professional competencies and skills

leaning towards Industry 4.0. In Proceedings of the 14th

International Conference on Computer Supported

Education - Volume 2: CSEDU (pp. 622–630).

SciTePress. https://doi.org/10.5220/0011047300003182

Naicker, N., Adeliyi, T., & Wing, J. (2020). Linear support

vector machines for prediction of student performance in

school-based education. Mathematical Problems in

Engineering, 2020, 1–7. https://doi.org/10.1155/2020/

4761468

Nourbakhsh, N., Wang, Y., Chen, F., & Calvo, R. A. (2012).

Using galvanic skin response for cognitive load

measurement in arithmetic and reading tasks. In

Proceedings of OZCHI 2012: 24–30 November 2012,

Melbourne, Victoria, Australia (pp. 420–429).

Melbourne: ACM. https://doi.org/10.1145/2414536.241

4602

Ofori, F., Maina, E., & Gitonga, R. (2020). Using machine

learning algorithms to predict students’ performance and

improve learning outcome: A literature-based review.

Journal of Information and Technology, 4(1), 23–45.

Pang, Y., Judd, N., O’Brien, J., & Ben-Avie, M. (2017).

Predicting students’ graduation outcomes through

support vector machines. In Proceedings of the IEEE

Frontiers in Education Conference (FIE) (pp. 1–8).

Indianapolis, IN, USA: IEEE. https://doi.org/10.1109/

FIE.2017.8190666

Pavani, M., Teja, A. R., Neelima, A., Bhavishya, G., &

Sukrutha, D. S. (2017). Prediction of student outcome in

educational sector by using decision tree. International

Journal for Technological Research in Engineering,

4(8), 1191–1193.

Pérez, F. de A., Santos-Gago, J. M., Caeiro-Rodríguez, M.,

& Fernández Iglesias, M. J. (2018). Evaluation of

commercial-off-the-shelf wrist wearables to estimate

stress on students. Journal of Visualized Experiments,

136, e57590. https://doi.org/10.3791/57590

Pijeira-Díaz, H. J., Drachsler, H., Kirschner, P. A., & Järvelä,

S. (2018). Profiling sympathetic arousal in a physics

course: How active are students? Journal of Computer

Assisted Learning, 34(5), 1–12. https://doi.org/10.1111/

jcal.12271

Poh, M. Z., Swenson, N. C., & Picard, R. W. (2010). A

wearable sensor for unobtrusive, long-term assessment of

electrodermal activity. IEEE Transactions on Biomedical

Engineering, 57(5), 1243–1252. https://doi.org/10.1109/

TBME.2009.2038487

Polyzou, A., & Karypis, G. (2023). Feature extraction for

classifying students based on their academic

performance. In Proceedings of the 11th International

Conference on Educational Data Mining (pp. 356–362).

Potter, L., Scallon, J., Swegle, D., Gould, T., & Okudan

Kremer, G. (2019). Establishing a link between

electrodermal activity and classroom engagement. In

IISE Annual Conference 2019 Proceedings (pp. 988–

993). Ames, IA: Industrial and Manufacturing Systems

Engineering, Iowa State University.

Quintero, H. F. P., & Bolkhovsky, J. B. (2019). Machine

learning models for the identification of cognitive tasks

using autonomic reactions from heart rate variability and

electrodermal activity. Behavioral Sciences.

https://doi.org/10.3390/bs9040045

Quintero, H. F. P., Florian, J. P., & Orjuela-Cañón, A. D.

(2016b). Highly sensitive index of sympathetic activity

based on time-frequency spectral analysis of

Enhancing Learning with Physiological Measures: A Systematic Review of Applications in Neuroeducation

121

electrodermal activity. American Journal of Physiology-

Regulatory, Integrative and Comparative Physiology,

311(3), R582–R591. https://doi.org/10.1152/ajpregu.00

180.2016

Quintero, H. F. P., Florian, J. P., Orjuela-Cañón, A. D.,

Aljama-Corrales, T., Charleston-Villalobos, S., & Chon,

K. H. (2016a). Power spectral density analysis of

electrodermal activity for sympathetic function

assessment. Annals of Biomedical Engineering, 44(10),

3124–3135. https://doi.org/10.1007/s10439-016-1606-6

Rastrollo-Guerrero, J. L., Gómez-Pulido, J. A., & Durán-

Domínguez, A. (2020). Analyzing and predicting

students’ performance by means of machine learning: A

review. Applied Sciences, 10(3), 1042.

https://doi.org/10.3390/app10031042

Reid, C., Keighrey, C., Murray, N., Dunbar, R., & Buckley,

J. (2020). A novel mixed methods approach to synthesize

EDA data with behavioral data to gain educational

insight. Sensors, 20(23), 6857. https://doi.org/10.3390/

s20236857

Rodic-Trmcic, B., Stanojevic, G., Sapic, R., Labus, A., &

Bogdanovic, Z. (2016). Wearable solution for assessing

physiological arousal towards students' interest and

engagement in the classroom. In Proceedings of the 11th

International Conference on Virtual Learning (pp. 215–

220). Belgrade: University of Belgrade.

Rodríguez-Arce, J., Lara-Flores, L., Portillo-Rodríguez, O.,

& Martínez-Méndez, R. (2020). Towards an anxiety and

stress recognition system for academic environments

based on physiological features. Computer Methods and

Programs in Biomedicine, 190, 105408. https://doi.org/

10.1016/j.cmpb.2020.105408

Sánchez-Carracedo, F., Trepat, E., & Barba-Vargas, A.

(2021). Successful engineering lecturing based on

neuroscience. International Journal of Engineering

Education, 37(1), 115–132.

Sánchez-Reolid, R., López, M. T., & Fernández-Caballero,

A. (2020). Machine learning for stress detection from

electrodermal activity: A scoping review. Preprints.

Albacete: Universidad de Castilla-La Mancha.

https://doi.org/10.20944/preprints202011.0043.v1

Schneider, B., Dich, Y., & Radu, I. (2020). Unpacking the

relationship between existing and new measures of

physiological synchrony and collaborative learning: A

mixed methods study. International Journal of

Computer-Supported Collaborative Learning, 15(2), 89–

113. https://doi.org/10.1007/s11412-020-09318-2

Shanthini, A., Vinodhini, G., & Chandrasekaran, R. M.

(2018). Predicting students' academic performance in the

university using meta decision tree classifiers. Journal of

Computer Science, 14(5), 654–662. https://doi.org/10.38

44/jcssp.2018.654.662

Simjanoska, M., Gusev, M., Ristov, S., & Bogdanova, A. M.

(2014). Intelligent student profiling for predicting e-

assessment outcomes. In Proceedings of IEEE Global

Engineering Education Conference (EDUCON 2014)

(pp. 616–622). Istanbul: IEEE. https://doi.org/10.1109/

EDUCON.2014.6826157

Terriault, P., Kozanitis, A., & Farand, P. (2021). Use of

electrodermal wristbands to measure students' cognitive

engagement in the classroom. In Proceedings of the

Canadian Engineering Education Association (CEEA-

ACEG21) Conference (pp. 1–8). UPEI. https://doi.org/

10.24908/pceea.vi0.14827

Thammasan, N., Stuldreher, I. V., Schreuders, E., Giletta, M.,

& Brouwer, A. M. (2020). A usability study of

physiological measurement in school using wearable

sensors. Sensors, 20(18), 5380. https://doi.org/10.3390/

s20185380

Villanueva, I., Campbell, B. D., Raikes, A. C., Jones, S. H.,

& Putney, L. G. (2018). A multimodal exploration of

engineering students' emotions and electrodermal

activity in design activities. Journal of Engineering

Education. https://doi.org/10.1002/jee.20225

Villanueva, I., Di Stefano, M., Gelles, L., Vicioso Osoria, P.,

& Benson, S. (2019). A race re-imagined, intersectional

approach to academic mentoring: Exploring the

perspectives and responses of womxn in science and

engineering research. Contemporary Educational

Psychology, 59. https://doi.org/10.1016/j.cedpsych.20

19.101786

Villarejo, M. V., Zapirain, B. G., & Zorrilla, A. M. (2012). A

stress sensor based on galvanic skin response (GSR)

controlled by ZigBee. Sensors, 12(5), 6075–6101.

https://doi.org/10.3390/s120506075

Wakelam, E., Jefferies, A., Davey, N., & Sun, Y. (2019). The

potential for student performance prediction in small

cohorts with minimal available attributes. British Journal

of Educational Technology. https://doi.org/10.1111/

bjet.12836

Walsh, K. R., & Mahesh, S. (2017). Exploratory study using

machine learning to make early predictions of student

outcomes. In Proceedings of the 23rd Americas

Conference on Information Systems (AMCIS 2017).

Boston: [s.n.].

Wang, Y., & Liu, J. (2019). A self-developed smart

wristband to monitor exercise intensity and safety in

physical education class. In Proceedings of the 8th

International Conference of Educational Innovation

through Technology (pp. 160–164). Biloxi, MS: IEEE.

https://doi.org/10.1109/EITT.2019.00038

Yan, L., & Liu, Y. (2020). An ensemble prediction model for

potential student recommendation using machine

learning. Symmetry, 12(5), 728. https://doi.org/

10.3390/sym12050728

Yoon, S., Sim, J. K., & Cho, Y.-H. (2016). A flexible and

wearable human stress monitoring patch. Scientific

Reports, 6, 23468. https://doi.org/10.1038/srep23468

Zabriskie, C., Yang, J., Devore, S., & Stewart, J. (2019).

Using machine learning to predict physics course

outcomes. Physical Review Physics Education Research,

15(2), 020120. https://doi.org/10.1103/PhysRevPhys

EducRes.15.020120

CSEDU 2025 - 17th International Conference on Computer Supported Education

122