ArticlePDF Available

Impact of Cognitive Fatigue on Attention and the Implications for Construction Safety: A Neuroscientific Perspective

June 2024
Journal of Construction Engineering and Management 150(8)

DOI:10.1061/JCEMD4.COENG-147

Authors:

Hongzhe Yue

Southeast University

Gui Ye

Chongqing University

Qinjun Liu

Western Sydney University

Show all 6 authorsHide

The global concern over fatigue in construction professionals, leading to diminished attention, unsafe behaviors, and even accidents, has escalated. Existing research on the impact of fatigue on attention has predominantly focused on qualitative behavioral aspects, relying extensively on self-reported measures and subjective assessments, resulting in interpretations with strong subjectivity and occasionally inconsistent findings. This study bridges this gap by adopting a neural perspective, utilizing a comprehensive quantitative measurement approach that integrates EEG (electroencephalograms), behavioral tests, and subjective rating scales. This interdisciplinary approach attempts to explore the neural mechanisms underlying the impact of fatigue on the attention of construction professionals, considering the regulatory effects of effort. Twenty participants from the construction sector were enlisted to undertake a 60-min Oddball cognitive task. The results indicate that as cognitive fatigue intensifies, the pattern of attention decline exhibits a slow-fast-slow trajectory. Initially, the dominance of effort is observed, which transitions to a stage where resource consumption takes precedence. In the later stage, participants tend to prioritize expediency over accuracy. The study synthesizes these outcomes to delve into the neural mechanisms of fatigue’s impact on attention, addressing the distinct phases, underlying mechanisms, and functions of attention. Moreover, it provides actionable recommendations to elevate attention levels and enhance safety in the construction industry, serving as a valuable guide for practical applications and further research in construction safety management.

Research framework: (a) participant recruitment; (b) experiment and data collection; and (c) data processing.

…

Experimental procedure: (a) process of experiment; (b) screen stimulate display; (c) data recording procedure; and (d) experimental site photo.

…

Fatigue and effort scale: (a) the Karolinska sleepiness scale; and (b) the rating scale for mental effort.

…

Electrode location.

…

Average reaction time of subjects in each stage.

…

Figures - uploaded by Qingting Xiang

Content may be subject to copyright.

Content uploaded by Qingting Xiang

Content may be subject to copyright.

Impact of Cognitive Fatigue on Attention and

the Implications for Construction Safety:

A Neuroscientific Perspective

Hongzhe Yue1; Gui Ye, Ph.D.2; Qinjun Liu, Ph.D.3;

Xiaohan Yang4; Qingting Xiang5; and Yalan Luo6

Abstract: The global concern over fatigue in construction professionals, leading to diminished attention, unsafe behaviors, and even

accidents, has escalated. Existing research on the impact of fatigue on attention has predominantly focused on qualitative behavioral aspects,

relying extensively on self-reported measures and subjective assessments, resulting in interpretations with strong subjectivity and occa-

sionally inconsistent findings. This study bridges this gap by adopting a neural perspective, utilizing a comprehensive quantitative meas-

urement approach that integrates EEG (electroencephalograms), behavioral tests, and subjective rating scales. This interdisciplinary

approach attempts to explore the neural mechanisms underlying the impact of fatigue on the attention of construction professionals, con-

sidering the regulatory effects of effort. Twenty participants from the construction sector were enlisted to undertake a 60-min Oddball

cognitive task. The results indicate that as cognitive fatigue intensifies, the pattern of attention decline exhibits a slow-fast-slow trajectory.

Initially, the dominance of effort is observed, which transitions to a stage where resource consumption takes precedence. In the later stage,

participants tend to prioritize expediency over accuracy. The study synthesizes these outcomes to delve into the neural mechanisms of

fatigue’s impact on attention, addressing the distinct phases, underlying mechanisms, and functions of attention. Moreover, it provides

actionable recommendations to elevate attention levels and enhance safety in the construction industry, serving as a valuable guide for

Society of Civil Engineers.

Practical Applications: This study underscores the effect of cognitive fatigue on the attention of construction professionals, directly

impacting on-site safety. With increasing fatigue, the attentiveness and response precision of these professionals diminish, escalating accident

risks. Implementing practical measures such as work schedule optimization, regular breaks, and fatigue monitoring through technology can

bolster safety. Additionally, site managers might consider task rotation and focus on cognitive resource management during training. These

strategies are designed to enhance both safety and productivity by prioritizing the cognitive well-being of construction professionals, guiding

companies in establishing more effective safety protocols in the workplace.

Author keywords: Construction professionals; Attention; Fatigue; Effort; Neuroscientific perspective; Electroencephalograms (EEG).

Introduction

The construction industry is notorious for its high accident rates

and subpar safety performance worldwide (Koc et al. 2023;

Namian et al. 2021;Zhang et al. 2023). These incidents not only

endanger the well-being of construction professionals but also

place significant financial strains on both employees and employ-

ers. Heinrich et al. (1950) pinpointed that human errors caused 88%

of all industrial accidents, with attention decline being a primary

factor (Ke et al. 2021;Reason 1990). In the complex and perilous

terrains of construction sites, construction professionals are prone

to overlook safety hazards, commit operational errors, or have be-

lated reactions when their attention diminishes, leading to severe

accidents (Ke et al. 2021;Khalid et al. 2021). Enhancing attention

levels can strengthen adherence to safety protocols and reduce the

likelihood of accidents (Endsley 1995). Hence, identifying the el-

ements that influence construction professionals’attention is cru-

cial for devising specific strategies to mitigate safety incidents

(Li et al. 2019).

Among the various factors affecting attention, fatigue stands out

as a significant determinant with a profound impact (Csatho et al.

2012;Lorist 2008;van der Linden and Eling 2006). Fatigue can be

described as a reduction in the capacity to carry out mental and/or

1Research Assistant, School of Management Science and Real Estate,

Chongqing Univ., Chongqing 400044, PR China. Email: 18209296728@

163.com

2Professor, School of Management Science and Real Estate, Chongqing

Univ., Chongqing 400044, PR China (corresponding author). Email:

yegui760404@126.com

3Associate Lecturer, School of Engineering, Design and Built Environ-

ment, Western Sydney Univ., Locked Bag 1797, Penrith, NSW 2751,

Australia. ORCID: https://orcid.org/0000-0002-4746-0480. Email: Qinjun

.Liu@westernsydney.edu.au

4Master’s Candidate, School of Management Science and Real Estate,

Chongqing Univ., Chongqing 400044, PR China. Email: 1073340473@qq

.com

5Ph.D. Candidate, School of Management Science and Real Estate,

Chongqing Univ., Chongqing 400044, PR China. ORCID: https://orcid

.org/0000-0001-8861-6245. Email: xiangqingting@yeah.net

6Master’s Candidate, School of Economics and Management,

Southwest Univ., Chongqing 400044, PR China. Email: 1132243153@

qq.com

Note. This manuscript was submitted on November 1, 2023; approved

on March 27, 2024; published online on June 13, 2024. Discussion period

open until November 13, 2024; separate discussions must be submitted for

individual papers. This paper is part of the Journal of Construction En-

gineering and Management, © ASCE, ISSN 0733-9364.

J. Constr. Eng. Manage., 2024, 150(8): 04024102

physical tasks effectively (Techera et al. 2018b). It can arise from

prolonged mental or physical exertion or as a result of a medical

condition (Kitani 2011). Particularly in the construction industry,

tasks of high intensity can easily lead to fatigue among construction

professionals, making them more susceptible to attentional lapses

and risk-taking behaviors (Techera et al. 2018a;Umer et al. 2021;

Wang et al. 2023;Xing et al. 2020). As a consequence, recognizing

the intricate relationship between fatigue and attention is pivotal

for enhancing safety measures and reducing hazardous incidents

among construction professionals.

However, there exists a notable lack of consensus regarding the

mechanisms through which fatigue impacts attention. the trajectory

of attention decline remains a topic of ongoing debate, with some

studies suggesting patterns such as an inverted U-shaped trend

(i.e., an initial increase followed by a decrease) or a consistent drop

(Faber et al. 2012;Larue et al. 2011). In addition, existing research

suggests that effort may serve as a compensatory mechanism to

regulate attentional decline during sustained tasks (Egeth 1975;

Westbrook and Braver 2015). However, in the construction indus-

try, the regulatory mechanisms by which effort affects fatigue on

attention remain unclear. Furthermore, the reasons, stages, and

functions of attention impacted by fatigue have not been fully

elucidated.

The disparities and ambiguities in existing literature might be

ascribed to a theoretical void in grasping the core neural mecha-

nisms, and the failure to effectively explain the results from a neu-

roscientific perspective. Current research methods heavily rely on

self-reports, subjective evaluations, and qualitative analyses (Chen

et al. 2018;Hasanzadeh et al. 2017). Consequently, interpretations

are largely restricted to the behavioral domain, leading to subjective

and inconsistent conclusions (Fu et al. 2022). While some endeav-

ors have been made to employ a quantitative approach in investi-

gating the relationship between fatigue and attention, they mainly

focus on refining measurement techniques rather than delving into

the neural underpinnings of attention decline caused by fatigue

(Li et al. 2019;Wang et al. 2023). A thorough exploration at the

neural level, leveraging EEG experiments methodologies that pri-

oritize objective quantitative analysis, can address these inconsis-

tencies. Shifting the research focus from the behavioral to neural

aspect can yield more robust evidence, thus paving the way for a

holistic understanding.

To bridge these research gaps, this study employs a primarily

quantitative EEG methodology for continuous cognitive tasks.

Behavioral and EEG data were systematically captured using

E-Prime 3.0 and an EEG recording system, addressing the limita-

tions of prior qualitative studies. Subjective scales assessed partic-

ipants’fatigue and effort levels, aiming to discern the compensatory

role of effort. By integrating experimental insights and relevant

theoretical frameworks, this paper delves into the impact patterns

of fatigue on attention regulation under the influence of effort and

elucidates the neural mechanisms underlying the effects of fatigue

on attention. The paper concludes by offering practical recommen-

dations to enhance attention levels and strengthen construction

safety.

Literature Review

Exploring the impact of fatigue on the safety performance of con-

struction professionals has largely been approached through quali-

tative methods. Namian et al. (2021) use of subjective fatigue

assessment scales revealed significant negative effects of fatigue

on hazard recognition and safety risk perception among construc-

tion workers, accounting for only 37% and 28% of the variance,

with other factors unconsidered. Conversely, Techera et al. (2018a)

identified the causes and consequences of fatigue through survey

questionnaires, suggesting that fatigue could lead to a slower work

pace and reduced attention based on the perspectives of transmis-

sion and distribution (T&D) workers. However, these findings are

based on workers’perspectives and not on objective facts.

In contrast, recent advancements have seen a shift toward quan-

titative approaches, with studies like Mehmood et al. (2023a) and

Wang et al. (2023) employing EEG to classify mental fatigue in

construction operators and workers into alert, mild fatigue, and

fatigue states. Such use of EEG offers an objective assessment by

measuring brain electrical activity, bypassing the subjective biases

of survey methods. However, these studies predominantly focus on

classifying fatigue states for monitoring purposes and do not elu-

cidate the neural mechanisms of fatigue’s impact on the behavioral

state of construction practitioners and the stages of impact.

While some experiments have investigated the mechanisms by

which fatigue influences attention using EEG, results have been

inconsistent—showing a decline trend (Larue et al. 2011) or an in-

verted U-shaped trend (Faber et al. 2012). It is noteworthy that their

subjects were not construction industry workers or employees, thus

not representative of the industry, and they did not consider the

regulatory mechanism of effort on attention, leading to incomplete

and potentially inaccurate insights into fatigue’s impact on atten-

tion. The relevant literature summary is shown in Table 1, there

is a gap in the literature regarding the neural mechanisms of fa-

tigue’s impact on attention under the moderating effect of effort

within the construction industry, which is the focus of this study.

Research Hypothesis

Continuous Cognitive Tasks Induce Construction

Professionals’Fatigue

Within the construction industry, fatigue primarily stems from

monotonous, repetitive tasks and the necessity for vigilance in

complex environments (Maynard et al. 2021). Continuous cogni-

tive tasks are particularly effective in catalyzing this fatigue.

Participants typically experience the time on task (TOT) effect in

these tasks, which is fatigue induced by the cumulative effect of

task duration. Engaging in cognitive tasks for prolonged periods,

often more than 30 min, has been shown to induce fatigue (Lorist

2008;Smith et al. 2016;Tanaka et al. 2009;Wang et al. 2023). This

phenomenon can be understood through the lens of the “resource

limitation theory.”Proposed by Egeth (1975), this theory suggests

that the cognitive processing system is constrained by a finite pool

of resources. When individuals engage in sustained cognitive tasks,

these resources are progressively depleted, leading to a decrease in

individual cognitive function. This depletion effect aligns with both

the perceptual and response selection model and the resource

depletion theory, emphasizing that sustained work consumption

taxes the cognitive system beyond its limits, inevitably resulting

in functional decline. Hence, this study proposes hypothesis H1.

Hypothesis H1: Conducting sustained cognitive tasks can in-

duce fatigue in participants.

The Role of Effort Regulation in Balancing Fatigue and

Cognitive Function in Task Performance

Individual attention levels and the exertion of effort are dynamic

processes that vary from moment to moment. The compensation

control theory, as articulated by Hockey (1997). Posits that individ-

uals actively regulate their cognitive and physiological resources

to compensate for any deficits incurred by fatigue. This theory

J. Constr. Eng. Manage., 2024, 150(8): 04024102

is particularly relevant in understanding how individual manage

their cognitive functions in the face of fatigue. In accordance with

this theory, during tasks that induce fatigue, participants are ob-

served to actively increase their effort to counterbalance the effects

of fatigue. This phenomenon is reflected in studies such as that by

Takacs et al. (2019), which found that moderate increases in fatigue

do not significantly impact cognitive function, possibly due to this

compensatory effort. Similarly, the cognitive system’s resistance to

high-demand task challenges can be attributed to participants’ac-

tive effort regulation. Hopstaken et al. (2015) further explored the

relationship between the degree of fatigue experienced by the par-

ticipants and their task participation, using metrics such as the rat-

ing scale for mental effort (RSME), behavioral indicators, and pupil

diameter indicators. Their findings revealed that as the task pro-

gressed, the score on the RSME scale increased significantly,

and the correlation between subjective effort and performance

indicators was stronger than the correlation between fatigue and

performance indicators, indicating that the participant’s task input

may be more dependent on effort regulation. Therefore, this study

proposes hypothesis H2.

Hypothesis H2: The level of effort made by the participants in-

creases as the continuous cognitive tasks progress.

The Impact of Fatigue on Attention: Behavioral and

EEG Evidence

Lim and Dinges (2008) explored the correlation between fatigue

induced by sleep deprivation and impairment of attention, sug-

gesting a decline in attentional resources under fatigue. This aligns

with the cognitive energy model proposed by Sanders (1983),

Table 1. Summary of literature on fatigue and its impact on performance in the construction industry and related fields

References Purpose Method Details

Namian et al. (2021) Assessing the impact of fatigue on

construction workers’safety

performance

Assessment scale method Fatigue significantly impacts workers’hazard

recognition and safety risk perception, but only explains

37% and 28% of the variance, not considering other

factors.

Techera et al. (2018a) Identifying and describing common

causes and consequences of fatigue

among Transmission and

Distribution (TD) workers

Survey questionnaire

method

Workers attribute extreme temperatures and long work

hours as main causes of fatigue leading to slower work

pace and reduced attention. However, findings are based

on workers’perspectives and not objective facts.

Ma et al. (2023) Monitoring fatigue levels of

construction workers using sweat

sensors

Sweat sensor measurement Sweat biomarkers can predict fatigue levels, but no direct

link between fatigue levels and attention or performance

of construction workers is established.

Zhang et al. (2023) Studying the impact of physical and

mental fatigue on unsafe behaviors

of construction workers

Simulated experiment

method, skin temperature

measurement

Physical and mental fatigue negatively affect cognitive

and motor abilities of construction workers. Mental

fatigue more likely changes workers’risk propensity

toward risk-taking. However, no explicit link between

fatigue and attention is mentioned.

Wang et al. (2023) Identifying mental fatigue in

construction workers

EEG and deep learning As mental fatigue increases, participants make more

cognitive errors in task completion, indicating a decrease

in workers’cognitive abilities and their ability to perform

tasks correctly. Miss rate used to represent task execution

capability, with no connection to neural mechanisms

explaining impact on workers’performance.

Fang et al. (2015) Studying the impact of fatigue on

construction workers’safety

performance

Manual handling task

simulation experiment

Fatigue level of 20 identified as a critical point where

fatigue effects begin to emerge. A linear relationship

exists between fatigue levels and error rates when fatigue

level exceeds 20. Safety performance assessed by

monitoring errors during task execution, no connection to

neural mechanisms explaining impact on workers’

performance.

Mehmood et al.

(2023b)

Classifying mental fatigue states in

construction equipment operators

Wearable EEG sensor data

and deep learning method

Mental fatigue states classified into alert, mild fatigue,

and fatigue states. The direct relationship between fatigue

levels and workers’performance or attention is not

explicitly discussed.

Larue et al. (2011) Assessing how road design and

roadside environment monotony

affect driver vigilance and driving

performance

EEG Participants included college students and others. The

study observed that the P300 amplitude decreased over

time following fatigue, demonstrating a general declining

trend. The regulatory role of effort was not considered.

Faber et al. (2012) Investigating how mental fatigue

affects visual selective attention

EEG Participants were 17 healthy volunteers. The study

observed a distinct U-shaped trend in participants’

responses, indicating that attention levels initially

increased, followed by a decrease, ultimately falling

below the initial level. The regulatory role of effort was

not considered.

J. Constr. Eng. Manage., 2024, 150(8): 04024102

which posits that cognitive processes are fueled by a finite energy

source. As tasks progress and fatigue sets in, this energy depletes,

leading to a decrease in cognitive performance such as attention.

Mean reaction time and accuracy, as crucial behavioral indicators

in cognitive experiments, are often used to reflect the cognitive

ability or information processing efficiency of participants (Wenger

and Gibson 2004). Mean reaction time represents the average time

taken by the participant to complete a phase of the task, while ac-

curacy is the proportion of correct responses made. Guo et al.

(2016) evaluated the damaging effects of fatigue on visual sustained

attention through experimentation and found that participants’re-

action speeds significantly slowed after fatigue, and their operation

accuracy also decreased, indicating a significant reduction in sus-

tained attention after fatigue. Similarly, Li et al. (2019) showed that

after fatigue induction, the detection rate of hazards decreases to

70% of the initial performance after 36 min of operation, and fur-

ther decreases to 60% after 60 min. As the tasks progress, their

reaction time also significantly increases.

Alternatively, P300 waves, identified in EEG as event-related

potentials (ERP), are critical markers in cognitive neuroscience

(Watford et al. 2020). The P300 is a positive wave in EEG record-

ings that occurs approximately 300 ms after the presentation of a

stimulus. It is widely used as an indicator of cognitive workload

and attentional resource allocation, making it a valuable tool in

studies examining cognitive functions. ERPs are electrical patterns

in brain activity that are elicited in response to specific sensory,

cognitive, or motor events (Spronk et al. 2008). The P300 compo-

nent of ERPs is particularly sensitive to the allocation of attention

resources and is often used as a reliable indicator of cognitive en-

gagement and attentional capacity (Käthner et al. 2014;Naito

et al. 2005). With their amplitude typically measured at key scalp

locations such as Cz—located over the midline of the scalp—and

Fz—situated above the forehead—these waves reflect the neural

activity associated with attentional demands. Decreased P300 am-

plitude at these electrode points is commonly linked with atten-

tional deficits, often resulting from fatigue. Zhao et al. (2012)

observed a significant reduction in P300 amplitude at Cz and Fz

following fatiguing tasks, signaling an impaired attentional func-

tion. Similarly, Boksem et al. (2005) documented a decrease in

P300 amplitude resulting from mental fatigue, highlighting the in-

timate relationship between fatigue-induced cognitive decline and

P300 amplitude. Thus, post-fatigue deterioration in attention is

manifested by alterations in behavioral data, such as increased

mean reaction time and reduced accuracy, alongside changes in

EEG data, notably the average P300 wave amplitude. This evidence

forms the basis for hypothesis H3.

Hypothesis H3: Compared to before fatigue, participants’

attention level significantly decreases.

Hypothesis H3-1: Compared to before fatigue, participants’

mean reaction time significantly extends.

Hypothesis H3-2: Compared to before fatigue, participants’

accuracy significantly decreases.

Hypothesis H3-3: Compared to before fatigue, participants’

average P300 wave amplitude significantly decreases.

According to the “maintaining rule of attention”in neuroscience

research (Hancock and Warm 1989), attention cannot be sustained

at the same intensity for more than half an hour, resulting in declin-

ing performance levels. As a limited resource (Egeth 1975), atten-

tion fluctuates in daily life, including in the context of work where

factors like fatigue affect attention levels. Individuals initially ex-

perience a period of heightened alertness and attention after

encountering a stressor, but as fatigue accumulates, they enter a

resistance period. To counteract the negative effects of fatigue and

sustain attention, individuals employ mechanisms such as effort,

arousal, and activation (Hockey 1997;Kurzban et al. 2013;Sanders

1983). Eventually, individuals reach a state of exhaustion charac-

terized by a significant decrease in attention. However, if physical

functioning or job capacity reaches a certain threshold of decline,

efforts to correct suboptimal arousal and activation will only pro-

vide temporary relief (Hockey 1997), suggesting that efforts may

not fundamentally alter the negative impact of fatigue on attention

but may attenuate the extent of the decline in certain stages affected

by fatigue, resulting in a nonlinear downward trend in attention.

Therefore, this paper proposes hypothesis H4.

Hypothesis H4: As the experiment continues, participants’

attention exhibits a nonlinear downward trend.

Methods

Fig. 1shows the overview of the experimental process. This study

began by recruiting eligible participants. After a rigorous screening

process, we ensured that only the most suitable candidates partici-

pated in the experiment, as depicted in Fig. 1(a). Fatigue was then

induced using the time on task (TOT) experimental procedures.

The subjective scale records the level of effort and fatigue, while

E-Prime 3.0 was employed for automated behavioral data record-

ing. Concurrently, an EEG acquisition system captured and re-

corded EEG data, as illustrated in Fig. 1(b). For data processing,

the EEG data underwent preprocessing using MATLAB. To ana-

lyze the different data sets effectively, we employed specific stat-

istical tests tailored to the nature of each data type, as illustrated in

Fig. 1(c). For the subjective data, a one-way repeated measures

ANOVA was used to assess within-subject effects over time, given

the repeated measures on the same individuals. This test is appro-

priate for analyzing changes in subjective measures like fatigue and

effort levels across different time points. For the behavioral data,

paired sample t-tests were chosen to compare the means of two

related groups (pre- and postfatigue stages) to understand how

fatigue impacts reaction time and accuracy. Lastly, using ANOVA

on EEG data to study scalp distribution and fatigue level interac-

tions helps us better understand how fatigue impacts various brain

areas and attention. All procedures outlined in Fig. 1received

approval from the Institutional Review Board of Chongqing

University.

Participants

This study recruited 20 eligible participants (6 males and 14

females) from the construction industry through a questionnaire

survey. These participants are early-career professionals within the

construction industry, predominantly aged between 18 to 25 years,

encompassing junior managers, supervisors, and construction in-

terns with at least one year of work experience. The selection cri-

teria for the participants included having normal corrected vision,

being right-handed, physically healthy, and self-assessing as hav-

ing good sleep quality. The higher number of female participants in

this study is attributable to their alignment with our recruitment

criteria and their greater willingness to participate. While the gen-

der distribution in our study may not be traditional, it reflects the

increasingly important role that women are playing in the con-

struction industry (Artis 2015). Notably, past studies have almost

exclusively involved male participants. Therefore, our study’sin-

clusion of female participants offers valuable insights into the cog-

nitive impacts within the evolving construction industry. To ensure

an unbiased sample, participants had no prior experience with this

type of experiment. Each participant provided informed consent

before the experiment and confirmed abstaining from stimulant

beverages in the 24 hours preceding the study. Additionally, a

J. Constr. Eng. Manage., 2024, 150(8): 04024102

preliminary questionnaire survey was conducted to determine

the most efficient and least fatigued times for the participants.

The experiments were then scheduled accordingly to coincide with

these peak alertness periods, maximizing the reliability and accu-

racy of our findings.

Experimental Procedure

The dual-stimulus Oddball paradigm, a well-established cognitive

task in neuroscientific research, was employed in this study to as-

sess attention and information processing. This paradigm involves

presenting two distinct types of stimuli to participants: standard

stimuli and target stimuli (Duncan et al. 2009;Strüber and Polich

2002). Standard stimuli appear frequently, establishing the norma-

tive context of the task, while target stimuli are less frequent, de-

signed to be noticeably different or unexpected in comparison to

the standard ones.

Prior research, such as Li et al. (2017), which focused on the

attention levels of miners, utilized wo (meaning I) and zhao (mean-

ing find) as standard and target stimuli, respectively. Similarly, Sun

et al. (2017) investigated the effects of rest on psychological fatigue

and cognitive performance using random letters like bas target

stimuli and + as standard stimuli. Our study draws from these

precedents, yet with a specific focus on the construction industry.

The task requires participants to respond selectively to the infre-

quent target stimuli, thus engaging key cognitive processes essen-

tial for attention and decision-making. The unique feature of the

dual-stimulus oddball task is its effectiveness in eliciting the P300

component, a significant event-related potential (ERP) in EEG

measurements, indicative of cognitive processing related to atten-

tion and decision-making (Strüber and Polich 2002).

In our study, we adapted the Oddball paradigm using the

Chinese characters wo and zhao to mimic the unpredictable and

dynamic scenarios often encountered in construction settings.

These characters were chosen based on their relevance to the cog-

nitive demands faced by construction professionals, who must rap-

idly adapt to unexpected events. wo symbolizes a standard state of

vigilance, corresponding to routine oversight without immediate

active engagement. Conversely, zhao represents sporadic, urgent

situations typical in construction sites, such as the sudden identi-

fication of safety hazards or the need to quickly locate essential

equipment. The infrequent appearance of zhao as the target stimu-

lus mirrors the rare but critical incidents in construction that require

immediate attention and action. This choice of stimuli aims to re-

duce variability from personal interpretations of specialized sym-

bols and enhance the relatability for our participant group, a

consideration supported by previous studies in similar high-risk

industries.

Fig. 1. Research framework: (a) participant recruitment; (b) experiment and data collection; and (c) data processing.

J. Constr. Eng. Manage., 2024, 150(8): 04024102

To effectively induce the P300 component, the oddball para-

digm in our study set the probability of target stimuli (i.e., zhao)

at 20%, with wo appearing as the standard stimulus 80% of the

time. This conforms to the recommended range of 10% to 20%

for target stimuli in the literature (Duncan et al. 2009). Moreover,

the task was structured to avoid presenting more than two consecu-

tive target stimuli.

Our experimental design included the arrangement of five

blocks in the 60-min oddball task, considering factors like presen-

tation time, method, and participant response. The presentation

time included various stages: instructions, fixation points, stimulus

display, intervals, and a concluding phase. Each fixation point, sig-

naling the participant to focus on the screen, lasted between 500

and 800 ms. The durations for stimulus display and intervals were

tailored to the specific needs of our experiment. Regarding presen-

tation methods, we combined automatic and key-triggered disap-

pearance of stimuli, which allowed for a more nuanced assessment

of response time and accuracy. Participants’responses were cap-

tured via a keyboard, facilitating precise measurement of reaction

times and accuracy.

Based on the previous content, the schematic diagram of the

experimental procedure is shown in Fig. 2(a).

Before the experiment, participants will see guidance prompts.

Once read, they can start the experiment by pressing any key.

Each trial begins with a fixation point + in the screen center,

disappearing after 500 ms. The stimulus then appears, disappearing

either after a key press or automatically after 1,500 ms if unac-

knowledged. A 500 ms blank screen follows before the next trial,

as shown in Fig. 2(b).

In each block, the nontarget wo appears 240 times, and the

target zhao appears 60 times. Participants are instructed to press

f for wo and j for zhao. Prior to the start of the first block and

following each subsequent block, participants are prompted to

complete a subjective rating scale before proceeding. To assess sub-

jective states of sleepiness and mental effort, this study utilized the

Karolinska Sleepiness Scale (KSS) and the Rating Scale for Mental

Effort (RSME), respectively.

he KSS [Fig. 3(a)], a nine-point scale ranging from 1 (extremely

alert) to 9 (very sleepy, great effort to keep awake), allows partic-

ipants to subjectively rate their level of sleepiness (Bonnefond et al.

2010). This scale is a validated tool commonly utilized in sleep

research to gauge subjective sleepiness and its variations during

different times of the day. The RSME [Fig. 3(b)] is used to measure

cognitive effort, with participants indicating the degree of their

effort on a scale from 0 to 150, reflecting an increasing level

of exertion (Hopstaken et al. 2015). This linear scale is a well-

established measure within cognitive and occupational research

fields, enabling the quantification of effort levels corresponding

to the mental workload experienced by participants.

During the experiment, both behavioral and EEG data are re-

corded. After the fifth block, an end prompt appears, directing par-

ticipants to the final rating scale, with data recording results in

Fig. 2(c). In total, accounting for questionnaire completion, the en-

tire procedure lasts around 60 min, with on-site photos shown in

Fig. 2(d).

Experiment Record

The experiments were conducted in a controlled environment, en-

suring optimal temperature, lighting, and soundproofing. During

the EEG paste application phase, casual conversations were initi-

ated to help participants remain relaxed. The impedance was con-

sistently monitored, ensuring it stayed below 10 kΩ. Before the

main experiment, a pre-recording phase was conducted to verify

the normality and stability of the EEG waveform. Any interference

detected prompted a check on the electrode resistance. Participants

were also guided through specific eye movements to confirm the

stability of the EEG waveform. Before the primary experiment, par-

ticipants underwent a familiarization phase to understand the re-

quirements. The experimenter ensured that no equipment would

Fig. 2. Experimental procedure: (a) process of experiment; (b) screen stimulate display; (c) data recording procedure; and (d) experimental site photo.

J. Constr. Eng. Manage., 2024, 150(8): 04024102

interfere with the experiment and verified the proper connection of

all devices. Participants were then positioned comfortably, focusing

on a computer display, while the experimenter monitored EEG

waveforms and keystroke markers to ensure participants’compre-

hension of the tasks.

Data Acquisition and Statistical Methods

Subjective Data

Throughout the experiment, participants were required to complete

six subjective scales, strategically placed between tasks. The data

collected was analyzed using SPSS 22.0, employing a single-factor

repeated-measures ANOVA. If the results did not meet the spheric-

ity assumptions, the Greenhouse-Geisser correction was applied.

Within this analytical framework, a significance level of P < 0.05

was set for statistical differences. When the P-value is less than

0.05, we can infer that the experimental conditions had a significant

impact on participants’levels of fatigue and mental effort. Fatigue

is considered a transitional stage between sleep and wakefulness

(Grandjean 1979). Fatigue and sleepiness are closely related, with

fatigue typically characterized by a need for more sleep and de-

creased alertness. Therefore, this study utilized the Karolinska

Sleepiness Scale (KSS) to assess fatigue levels (Akerstedt and

Gillberg 1990;Bonnefond et al. 2010), as shown in Fig. 3(a). Con-

currently, the RSME to measure participants’mental effort (Zijlstra

1993), as shown in Fig. 3(b).

Behavioral Data

The experimental tasks were presented using E-prime 3.0 software.

Before the formal experiment began, participants were required to

undergo a practice session to eliminate the practice effect, and data

from this session were not included in the behavioral data analysis.

After the formal experiment began, the procedure involved both a

stimulus and a blank interface, with both allowing for button re-

sponses. The software recorded response times, excluding missed

responses and noting any errors. E-prime 3.0 documented both re-

sponse time and accuracy, with the data subsequently processed in

SPSS 22.0.

EEG Data

EEG data was recorded using the actiCHamp brain recording and

analysis system (Brain Products), with stimuli presented using

E-prime 3.0 and EEG recorded using Vision Recorder 2.0. Elec-

trode placement followed the international 10–20 extended system

(Nuwer 1998), using a 64-channel electrode cap, as shown in Fig. 4.

The 10 and 20 refer to the actual distances between adjacent elec-

trodes, either 10% or 20% of the total front–back or right–left dis-

tance of the skull. Each electrode placement is identified using a

combination of letters and numbers. The letters refer to the brain

region (e.g., F for frontal, T for temporal, P for parietal, O for

occipital, and C for central), while the numbers or odd/even des-

ignations indicate the hemisphere location (odd numbers for the left

hemisphere, even numbers for the right hemisphere, and z for

midline positions). In our study, electrode placement followed this

10–20 system to ensure accurate and reliable recording of brain

activity across participants, particularly focusing on those regions

of the brain that are associated with attentional processes.

The ground electrode was placed at Fpz, and the online sam-

pling rate was 1,000 Hz. Before the experiment, electrode imped-

ances were reduced to below 10 kΩto ensure optimal signal

quality. Impedance refers to the resistance encountered by the elec-

trical current as it travels from the electrodes on the scalp through

the skin, tissue, and bone to the electrical circuits of the EEG re-

cording system. Lower impedance levels, typically below 10 kΩ,

are targeted to minimize signal attenuation and noise interference,

Fig. 3. Fatigue and effort scale: (a) the Karolinska sleepiness scale; and (b) the rating scale for mental effort.

Fig. 4. Electrode location.

J. Constr. Eng. Manage., 2024, 150(8): 04024102

which can be caused by poor electrode contact or high resistance in

the electrode-skin interface. This 10 kΩthreshold was chosen

based on empirical evidence (Keil et al. 2014;Laszlo et al. 2014).

EEG data preprocessing was performed using eeglab, a MATLAB-

based software (Delorme and Makeig 2004).

The steps included re-referencing, filtering, segmenting, base-

line correction, interpolating of bad channels and removing bad

segments, artifact rejection, averaging, etc. This study used a band-

pass filter (0.05–40 Hz) and a notch filter (48–52 Hz) to filter out

50 Hz electrical interference. The EEG capture time was from

200 ms before the presentation of the stimulus material to 800 ms

after the presentation, with the 200 ms before the presentation serv-

ing as the baseline. Data with an amplitude greater than 80 μV

were removed as other artifacts.

A two-factor repeated-measures ANOVA was conducted for

scalp distribution (frontal, central, and parietal regions) ×fatigue

(before and after), with individual effects analyzed when an inter-

action effect was present (P < 0.05). The EEG data was processed

to improve the signal-to-noise ratio (SNR) and remove artifacts us-

ing independent component analysis (ICA). The same preprocess-

ing procedures were applied to each participant’s EEG data, with

tracking of the preprocessing steps.

Results

Subjective Data Analysis

As the duration of the cognitive task extended, participants expe-

rienced varying degrees of fatigue. Participants were instructed to

complete the KSS scale both before the task’s commencement

(block 0) and following the conclusion of each block. The box

plots and line graphs in Fig. 5depict the fatigue levels of the

20 participants across six time intervals. A clear accumulation

of fatigue is observed as the task progresses. Postexperiment feed-

back from participants highlighted pronounced feelings of sleepi-

ness, irritability, and restlessness throughout the task. A one-way

within-subjects ANOVA was conducted on the fatigue levels

across the six intervals. After applying the Greenhouse-Geisser

correction, the results yielded Fð2.263;42.998Þ¼43.554,P<0.05.

After applying the Greenhouse-Geisser correction, the results

yielded Fð2.263;42.998Þ¼43.554,P<0.05. This signifies that

there are significant differences in fatigue levels over time. Sub-

sequent multiple comparisons revealed significant differences in

fatigue levels between each block, as detailed in Table 2. Notably,

there was a significant increase in fatigue levels post-task com-

pared to pretask, corroborating hypothesis H1.

With the deepening of fatigue, participants’operational ability

will be affected. To maintain task performance and maintain the

stability of operational resources, participants will continuously

improve their efforts to actively change their state. In this paper,

participants were required to fill in the RSME scale before the task

(block 0) and after each block. Calculate the box plots and line

graphs of the RSME scale for effort levels for 20 participants

at 6-time points, as shown in Fig. 6. Additionally, a one-way

repeated measures analysis of variance was conducted to ex-

amine the level of effort at six time points. After applying the

Greenhouse-Geisser correction, the results revealed a significant

effect, Fð2.426;46.101Þ¼58.288,p<0.05. Multiple compari-

sons showed significant differences in effort level between each

block, except for block 0 and block 1, as shown in Table 3.Com-

bining the results from Fig. 6and Table 3, It can be seen that with

the progress of the task, the effort degree of the participants

showed an overall upward trend, satisfying hypothesis H2.

Behavioral Data Analysis

The mean reaction times for the 20 participants across the five

blocks were calculated and presented in Fig. 7. The mean reaction

Fig. 5. Box plots and line graphs of the fatigue level. Fig. 6. Box plots and line graphs of the effort level.

Table 2. Multiple comparisons of different fatigue stages of the

participants

First

compared

block

Second

compared

block

Mean

difference

Standard

error Significance

block0 block1 −0.600 0.222 0.014

block2 −1.150 0.284 0.001

block3 −1.700 0.349 0.000

block4 −2.350 0.342 0.000

block5 −3.200 0.367 0.000

block1 block2 −0.550 0.135 0.001

block3 −1.100 0.191 0.000

block4 −1.750 0.239 0.000

block5 −2.600 0.294 0.000

block2 block3 −0.550 0.153 0.002

block4 −1.200 0.225 0.000

block5 −2.050 0.276 0.000

block3 block4 −0.650 0.150 0.000

block5 −1.500 0.212 0.000

block4 block5 −0.850 0.150 0.000

J. Constr. Eng. Manage., 2024, 150(8): 04024102

time reflects the participants’keying speed. As fatigue increased,

the participants’reaction times to stimulus presentation gradually

slowed down, but in the final stage of the experiment, it suddenly

became faster, possibly due to self-motivation toward the end of the

experiment. Pairwise t-tests were conducted on the mean reaction

times for each participant using block 1 as a prefatigue stage and

block 5 as a postfatigue stage. The results showed t ¼−2.153,

P<0.05. It can be seen that the mean reaction time of the partic-

ipants after fatigue was significantly prolonged compared to before

fatigue, indicating a significant decrease in attention, consistent

with hypothesis H3-1.

The mean accuracy of 20 participants in five blocks was calcu-

lated as shown in Fig. 8. Accuracy represents the participants’re-

sponse accuracy to target stimuli, which gradually decreased as the

task progressed. block 1 was used as the prefatigue stage, and block

5 was used as the postfatigue stage. A paired t-test was performed

on the accuracy of the participants before and after fatigue, with the

result of t ¼4.372,P <0.05. It can be seen that compared to before

fatigue, the accuracy of the participants after fatiguing significantly

reduced, indicating a significant decline in attention, which satisfies

hypothesis H3-2.

EEG Data Analysis

Differences in Attention Levels before and after Fatigue

Through the superimposed processing of EEG data, the P300

waveforms induced by the experiment can be observed. Based

on the characteristics of the overall average waveform and the dis-

tribution and meaning of the P300 component, nine electrodes, in-

cluding F3, Fz, F4, C3, Cz, C4, P3, Pz, and P4, were selected for

analysis. Then, the average waveforms before and after the fatigue

of the typical electrodes Fz, Cz, and Pz were plotted, using the ap-

pearance of a target stimulus as a lock time and 200ms before the

target stimulus as the baseline, as shown in Fig. 9.

Usually, the maximum positive waveform appearing within the

range of 280–550 ms is identified as the P300 wave (Naito et al.

2005). From the figure, it can be seen that the peak of the P300

appears at around 480ms, and the time window of 440–520 ms

was selected as the period of analysis, as indicated by the gray area,

which represents a clear P300 component. Using Block 1 as the

pre-fatigue stage and Block 5 as the post-fatigue stage, the average

P300 amplitudes of the nine analyzed electrodes were derived

and shown in Table 4. It can be seen that compared to before

fatigue, the average P300 amplitude of each electrode decreased

after fatigue.

To better explore the underlying mechanisms, this study di-

vided the electrode distribution area into three levels (frontal lobe:

F3, Fz, F4; central region: C3, Cz, C4; parietal lobe: P3, Pz, P4)

and categorized fatigue into two levels (pre-fatigue: block 1; post-

fatigue: block 5), conducting a 3×2repeated measures analysis of

variance. The statistical results, after Greenhouse-Geisser correc-

tion, showed a significant interaction effect between scalp distri-

bution and fatigue (Fð1;21Þ¼12.118,p<0.05), indicating the

need for further analysis of the separate effects of within-group

factors.

For the separate effect analysis of scalp distribution factors,

the performance of the P300 component in scalp distribution is

consistent between pre-fatigue and post-fatigue conditions. The

scalp distribution of average P300 amplitude before fatigue

showed a pattern of central region >parietal lobe >frontal lobe

(F ¼52.172,p<0.05), while the scalp distribution of average

P300 amplitude after fatigue showed a pattern of central region

>frontal lobe >parietal lobe (F ¼23.872,p<0.05). Analyzing

the P300 amplitudes of the frontal lobe, central region, and pari-

etal lobe before and after fatigue, as shown in Table 5, the P300

amplitudes significantly decrease in the central region (13.73 ver-

sus 11.15 μV), parietal lobe (11.78 versus 8.33 μV), and frontal

Table 3. Multiple comparisons of different effort stages of the participants

First

compared

block

Second

compared

block

Mean

difference

Standard

error Significance

block0 block1 −2.500 1.230 0.056

block2 −13.000 1.792 0.000

block3 −17.000 2.417 0.000

block4 −23.000 2.626 0.000

block5 −31.500 2.741 0.000

block1 block2 −10.500 1.698 0.000

block3 −14.500 2.348 0.000

block4 −20.500 2.854 0.000

block5 −29.000 2.982 0.000

block2 block3 −4.000 1.522 0.017

block4 −10.000 2.294 0.000

block5 −18.500 2.542 0.000

block3 block4 −6.000 1.835 0.004

block5 −14.500 2.112 0.000

block4 block5 −8.500 1.500 0.000

Fig. 7. Average reaction time of subjects in each stage.

Fig. 8. Accuracy of subjects in each stage.

J. Constr. Eng. Manage., 2024, 150(8): 04024102

lobe (11.18 versus 9.12 μV) after fatigue (p < 0.05), supporting

hypothesis H3-3.

Patterns of Changes in Attention Levels

Based on the experimental data, it was found that compared to pre-

fatigue, participants exhibited significantly longer reaction times,

lower accuracy, and reduced average P300 amplitude after fatigue.

Since EEG data provide more accurate measurements, this study

will focus on analyzing changes in attention levels using EEG

data, while the results from behavioral data can be used as a refer-

ence in the discussion. The changes in attention levels, fatigue lev-

els, and effort levels, represented by the absolute value of the slope

K, are shown in Table 6. There were significant differences in

jKAttentionjvalues between each block, indicating a nonlinear

Fig. 9. Total average waveform of Fz, Cz, and Pz electrodes.

Table 4. P300 average amplitude at each electrode (μV)

Stage

Electrode

F3 Fz F4 C3 Cz C4 P3 Pz P4

Pre-fatigue 10.12 11.90 11.54 12.39 15.08 13.73 10.62 13.46 11.24

Post-fatigue 8.24 9.62 9.50 10.52 12.38 10.56 7.76 9.66 7.56

Table 5. Separate effect analysis of fatigue factors (μV)

Scalp distribution Electrodes Pre-fatigue Post-fatigue Difference F P

Frontal lobe F3, Fz, F4 11.18 5.60 9.12 4.73 2.06 4.486 0.048

Central area C3, Cz, C4 13.73 6.34 11.15 5.18 2.58 9.398 0.006

Parietal lobe P3, Pz, P4 11.78 6.26 8.33 5.08 3.45 18.408 0.000

Note: The pre-fatigue and post-fatigue wave amplitudes are the averages of three electrodes selected from the frontal lobe, central area, and parietal lobe, and

the difference is the amplitude difference between post-fatigue and pre-fatigue.

J. Constr. Eng. Manage., 2024, 150(8): 04024102

decline in attention levels throughout the experiment. The overall

pattern showed a slow decline, followed by a faster decline, and

then a slower decline again, supporting Hypothesis H4.

Discussion

The Dynamic Process of Fatigue’s Impact on Attention

under Effortful Regulation

Traditionally, research in the construction sector has predominantly

relied on qualitative methods like self-reports to measure the rela-

tionship between attention and fatigue. In contrast, our study

employs a comprehensive quantitative approach, encompassing

subjective data, behavioral experiments, and EEG tests. This meth-

odological transition provides a clearer picture of participants’

physiological and psychological conditions, thereby circumventing

the inherent ambiguities in subjective self-reports.

Supported by both subjective and EEG data, our findings

indicate that prolonged cognitive tasks induce fatigue, which sub-

sequently impairs attention. This is manifested in elongated reac-

tion times and diminished accuracy, which is consistent with prior

research (Gergelyfi et al. 2015;Lim and Dinges 2008). Further-

more, our EEG data reveals a decline in P300 amplitude following

fatigue, corroborating with Boksem et al. (2005).

Upon further examination of the changing trends in subjective

data, behavioral data, and EEG data, it is observed that while there

is some variation in the degree of fatigue and effort perception

among participants, all participants display a fundamentally consis-

tent pattern of perception changes over time, exhibiting an upward

trend under the same experimental conditions. The overall accuracy

of the participants shows a downward trend, while the average re-

sponse time exhibits an upward trend, albeit decreasing after enter-

ing the fifth stage, indicating an acceleration in the participants’

response speed. The average P300 amplitude of the participants

follows a direct descending trend, with the amplitude differing

at each stage, aligning with the research results of Larue et al.

(2011) on the change patterns of attention. The downward pattern

of attention in this article is characterized by a slow, fast, and

slow trend.

Drawing on Resource limitation theory (Egeth 1975), compen-

sation control theory (Hockey 1997), cognitive energy model

(Sanders 1983), and the observation of fatigue and effort levels in

each stage, the following patterns of attention decline can be dis-

cerned: at the onset of the experiment, the participants’fatigue level

decreases without significant effort, indicating that the psychologi-

cal state and effort budget is at a relatively optimal level, meeting

the extra resource needs. The decline amplitude of attention from

Block 1 to Block 2 is relatively slow, potentially due to the partic-

ipants’efforts to shift from the initial to higher levels, increasing the

upper limit of effort expenditure, thereby attenuating resource con-

sumption and causing attention decline not to be significant. The

decline amplitude of attention from Block 2 to Block 3 is the

largest, suggesting that the participants’increased effort level may

not be sufficient to support cognitive resource recovery, resulting in

a significant decline in attention despite their increased effort. The

decline amplitude of attention from Block 3 to Block 5 begins to

slow down, but the final attention level of the participants is still far

lower than the initial stage. According to the post-experiment inter-

view, the participants had a certain perception of time during the

entire experiment. By the time they enter Block 4, they are aware

that the experiment is nearing its end and gradually start self-

motivating, leading to a slower decline in attention. In block 5, most

participants expressed a desire to end the experiment as soon as

possible, and the trade-off between speed and accuracy tends to

favor the former. This motivation to finish quickly to some extent

slows down the decline in attention level, which is also corrobo-

rated by changes in behavioral data.

In summary, the impact of fatigue on attention involves not only

the consumption of resources due to fatigue but also the compen-

sation of resources through effort. As effort can be seen as the will-

ingness and ability to actively change one’s psychological state,

observing attention changes from a dynamic perspective reveals

that the entire decline process is not linear and allows for a more

comprehensive interpretation of the dynamic process of fatigue’s

influence on attention, as shown in Fig. 10. This process can be

understood through three stages: early, middle, and late. During

the early stage of work, fatigue causes a decline in attention resour-

ces but is still sufficient to meet task demands. Individuals can en-

hance their mental activity energy through active coping, with

subjective effort regulation playing a dominant role. In the middle

stage, as fatigue deepens, individuals enter a passive coping mode.

The effort level may further increase to adapt to the new demand

level, but with limited effectiveness, as objective resource con-

sumption becomes dominant. In the late stage, subjective scales

show an increase in the effort level of the participants, but it may

only be a willingness to exert effort. According to the resource limi-

tation theory, the effort may have reached its limit. In this study,

participants choose to speed up or neglect accuracy to finish the

experiment as soon as possible. In this stage, objective resource

consumption construction speed becomes even more dominant.

In theoretical terms, this study confirms the hypothesis proposed

by Hopstaken et al. (2015) that attention levels also depend on sub-

jective effort regulation. Specifically, the resource limitation theory

focuses mainly on the consumption of attention resources but does

not fully explain the effort regulation provided by current willing-

ness (Egeth 1975). The compensation control theory and cognitive

energy model focus on the regulating role of effort but do not fully

explain the consequences of resource consumption (Hockey 1997;

Sanders 1983). Therefore, the nonlinear decline of attention sug-

gests that resources and effort alternately occupy a dominant role

in different stages of attention change. This reveals the core view-

points of the three major theories at the neural level, enhancing their

explanatory power. Since the decline in attention can be divided

into three stages (early, middle, and late), the practical implications

include activating construction professionals’awareness of effort in

Table 6. Comparison of changes in fatigue, effort, and attention level

Property block0 block1 block2 block3 block4 block5

Degree of fatigue (scores) 3.35 3.95 4.50 5.05 5.70 6.55

Level of effort (scores) 51.00 53.50 64.00 68.00 74.00 82.50

Attention level (μV) —12.23 12.01 10.36 9.97 9.53

jKFatiguej—0.18 0.14 0.12 0.13 0.15

jKEffortj—0.05 0.20 0.06 0.09 0.11

jKAttentionj——0.02 0.14 0.04 0.04

J. Constr. Eng. Manage., 2024, 150(8): 04024102

the early stage, reasonably arranging rest time for construction

professionals in the middle stage, and considering increasing mo-

tivation incentives in the late stage. Implementing these measures

not only reduces accident rates in the construction industry but also

provides a reference for safety strategy development in other fields,

contributing to improving safety management across various

sectors.

Neurophysiological Mechanisms Underlying the Impact

of Fatigue on Attention

Combining experimental results, the resource limitation theory, and

the brain function system theory were analyzed to examine poten-

tial neural mechanisms that may be affected by fatigue on attention,

as illustrated in Fig. 11.

At the onset of the experiment, pronounced P300 components

were observed in the frontal, central, and parietal regions, indicat-

ing sufficient attentional resource allocation. However, postfatigue,

a significant reduction in the P300 components across these regions

was noted, suggesting that fatigue progressively leads to attentional

decline, as depicted in Module A of Fig. 11.

Analyzing the reasons for the decline in attention due to fatigue,

some participants became distracted during the experiment, leading

to prolonged reaction times. This might be attributed to a reduced

total amount of attention resources allocated to the experiment. Ac-

cording to the Resource limitation theory (Egeth 1975), escalating

fatigue diminishes the total attention dedicated to the experiment,

thereby influencing attention functions, as illustrated in Module B.

Specifically, two pathways emerge: One factor is the decrease in the

total amount of resources, which leads to a decrease in the total

amount of available resources that can be activated, subsequently

affecting the allocation of resources and ultimately impacting atten-

tion, as shown in Route 1 of Module B in Fig. 11. The other path-

way suggests that as fatigue intensifies, construction professionals

become increasingly distracted, and in the latter stages of the ex-

periment, they may even desire to expedite its conclusion. The cur-

rent willingness influences the strategy of resource allocation,

thereby affecting the final allocation of attentional resources, as

shown in Route 2 of Module B in Fig. 11.

Conversely, while the accuracy rate of participants continued to

diminish, there was a resurgence in reaction time during the con-

cluding stage. This might be because participants allocated more

resources in the final stage to prioritize speed. However, due to

the neural activity impairment induced by fatigue, the accuracy rate

still deteriorated despite the augmented resource allocation. This

perspective is echoed by the brain function system theory (Luria

1973). The theory posits that fatigue’s impact on attention might

stem from impairments in certain neural activities in the pari-

etal-temporal-occipital association cortex and the limbic associa-

tion cortex area. This weakens the information transmission

between these regions, leading to a fading of attention functions,

as depicted in Module C of Fig. 11. Consequently, the process of

fatigue impairment encompasses both a reduction in the total

amount of allocatable attention resources and damage to neural ac-

tivity. Therefore, interventions should target both the reduction in

overall attention resources and the neural activity impairment to

effectively counteract the decline in construction professionals’

attention.

Whether it is a fatigue-induced reduction in overall attention

resources or fatigue impairing specific neural activities, the impact

of fatigue on attention primarily affects functions such as selection,

maintenance, and allocation of attention (Gomes et al. 2000).

Early stage of

the task

Middle stage

of the task

Late stage of

the task

Attention

Amount of attention

resources

Task demand resources

meet

level of effort

gap

Lower-level

regulation

Amount of attention

resources

Task demand

resources

miss

level of effort

gap

Higher-level

regulation

Effective

Amount of attention

resources

Task demand resources

miss

level of effort upper

limit

gap

Higher-level

regulation

Inefficient

Rapid decline Slow decline

Slow decline

Allocation Allocation Allocation

Proactive response safe

operations Passive response cause a safety accident

Attention Attention

Fatigue leads to a decrease in attention level and individual willingness, as well as a decline in the total amount of resources

and

individual devoted attention resources available for goal-directed tasks

Activating construction

professionals’ awareness of effort

Arranging rest time for

workers

Increasing motivation

incentives

Management

measures

Management

measures

Fig. 10. Dynamic process of fatigue affecting attention under the moderating effect of effort.

J. Constr. Eng. Manage., 2024, 150(8): 04024102

Participants needed to react differently to various stimuli to meet

the current task requirements, while simultaneously suppressing

irrelevant internal activities to avoid affecting task performance, in-

dicating that the fundamental function of attention is selection.

Throughout the experiment, participants were able to respond to

different stimuli, and since there were not many distractions in this

experiment, the function of attention selection remained effective.

Second, due to the continuous nature of the task and the random-

ness of the target stimuli, participants needed to maintain constant

attention to the input of information, as any lapse could result in

missing the appearance of the target stimuli, indicating the function

of attention maintenance. In the experiment, especially toward the

latter stages, participants missed target stimuli and pressed empty

keys, suggesting that the function of attention maintenance may

have been affected. Additionally, the appearance of target stimuli

was random and of low probability, so individuals needed to

promptly allocate resources to unexpected events during the experi-

ment to ensure that activities progressed toward anticipated goals

and requirements, indicating the function of attention allocation. In

the experiment, if the function of attention allocation was insuffi-

cient, construction professionals might exhibit slow or erroneous

key presses, which are manifestations of the inadequate allocation

function. Therefore, recommendations to enhance attention levels

should be made considering these three functions of attention.

Practical Case Study

In order to integrate our theoretical findings into practical applica-

tions, we investigated a real-world project scenario, focusing on the

task of a construction worker or a junior manager responsible for

site layout—a critical job requiring high precision to mark locations

on the ground where structures will be erected. The process starts in

the morning (early stage) with the worker or manager fresh and

attentive, reviewing site plans and preparing layout tools. The cog-

nitive load is manageable at this point, with attentional resources

optimal, as indicated by the higher P300 amplitude in the EEG data

(Fig. 11). Although there is a slight decline in attentional resources

at this stage, the decrease is still sufficient to meet the demands of

the task, as shown in the first part of Fig. 10.

As the day progresses toward noon (middle stage), the profes-

sionals’fatigue sets in due to prolonged labor, replacing the earlier

effect of rising temperatures with the onset of cognitive fatigue.

Frontal lobe

Average response

time increases and

accuracy decreases

Decrease in the average P300

amplitude across different brain

regions after fatigue.

Central region

Parietal lob e

Encoding and transmission of

input informati on by the brain

the prefrontal

association cortex

the parieto-

temporo-occipital

association cortex

the limbic association

cortex area

Sensory cortes

Recep tors

Motor cortex

Effectors

The maintenance function of attention

Observati on

of P300

componen t

Result in

Fatigu e affects brain and nervou s system functions

Arousal Avaliable

capacity

Final

alloca tion of

resour ces

Ongoi ng activit ies

and response

The total number

of resources

Fatigue leads to a decrease in the total amount of attent ional resources

Attention d ecreases

Electroence

phalo gram

(EEG) data

Core

expla nation

of attention

decrease

Manifestat

ions of

decreased

attent ion

function

Module C

the Brain

functi on

system theo ry

Module B

the Resourc e

limitation th eor y

Result in

Fatigue stimulus

Route 1

Resource allocation

schemes

Current

willingness

Route 2

Module B Module C

Module A

Module D

Impaired neuronal ac tivity

Dama ge Dama ge

The selective functio n of attentio n

The allocating function of attention

Result in

The dynami c

modulatory role

of effort

Fig. 11. Neural mechanism of the effect of fatigue on attention.

J. Constr. Eng. Manage., 2024, 150(8): 04024102

The precision required for marking out foundations demands sus-

tained attention, which begins to waver under the increasing cog-

nitive load. This is captured by the elongated reaction times and

decreased accuracy in behavioral experiments. The total number

of resources and arousal available capacity begin to decline, as de-

picted in Module B of Fig. 11. At the same time, the amount of

attention resources reaches the task demand resources, with con-

struction professionals still increasing their effort to complete

the task diligently, as illustrated in the second stage of Fig. 10.

Despite their efforts, the manager or worker may start to overlook

small details or make minor errors in measurements.

By late afternoon (late stage), the individual has been working

for several hours. The significant fatigue impacts their attention to

detail, and they may rush to complete the job, potentially leading to

mistakes in the site layout with serious repercussions for the con-

struction project. This period marks the peak effort to maintain at-

tention, as the willingness to continue exerting effort has already

reached the upper limit of the natural resource limitations of cog-

nitive endurance, as shown in the third stage of Fig. 10.

Based on the summarized theoretical models in Figs. 10 and 11,

we propose the following strategies to mitigate the impact of cog-

nitive fatigue on site layout tasks:

1. Implement safety and task mobilization on-site to stimulate

construction professionals’effort, responding to the activating

construction professionals’awareness of effort suggestion in

the first stage of Fig. 10.

2. Scheduled breaks: Incorporate short, frequent breaks into the

site layout process, particularly during the middle stage when

fatigue starts impairing attention. This aligns with the initial de-

cline in P300 amplitude and helps restore cognitive resources.

3. Team rotation: Rotate tasks among team members to reduce the

continuous cognitive load on any single individual, allowing for

a recovery period which helps maintain attention.

4. Hydration and cooling: It is essential to ensure that construction

professionals maintain proper hydration and thermal regulation,

potentially facilitated by shaded rest areas, to counteract the

physiological contributors to cognitive fatigue.

5. End-of-day review: Implement an end-of-day review process to

check the accuracy of layout work. This could include cross-

checking with another team member to mitigate the risk of er-

rors due to the decline in attentional resources.

6. Immediate positive feedback: Establish a routine of providing

immediate positive feedback to acknowledge the efforts of con-

struction professionals.

Strategies 2, 3, and 4 respond to arranging rest time for construc-

tion professionals in the second stage of Fig. 10, i.e., using breaks

to check for errors in tasks and reducing potential hazards due to

fatigue-related oversights. Strategies 5 and 6 respond to the need

for increasing motivational incentives in the third stage of Fig. 10,

aiming to stimulate construction professionals. Strategy 5 involves

end-of-day team reviews to acknowledge and encourage ongoing

diligence in their tasks, while Strategy 6 focuses on providing im-

mediate positive feedback.

Prior to our study, project managers seldom implemented mea-

sures to reduce cognitive fatigue, with only occasional checks and

safety reminders. Our recommendations have been adopted by the

managers, who hope to improve on-site safety and boost safety per-

formance through these measures.

Limitations and Future Research

It is crucial to acknowledge the limitations of this study. While our

research predominantly utilized EEG as a quantitative method to

investigate the neural mechanisms underlying the influence of

fatigue on attention, it is worth noting that attention itself can

be affected by multiple factors including lighting, noise, tempera-

ture, and the features of safety signs. Beyond the reasons, attention

allocation functions, and impact stages discussed in this paper,

other dimensions—such as employee’s age, gender, and job type,

might also influence attention decline. The potential impact of these

factors on attention could serve as directions for future research.

One limitation of our study is related to the experiment’s setup

and the nature of the visual stimuli used. The Oddball paradigm,

employing static Chinese characters as stimuli, contrasts sharply

with the dynamic, complex environment of a construction site.

In real-world situations, construction professionals face diverse vis-

ual and sensory cues, such as moving objects, varying lighting, and

a range of colors and shapes, which are not replicated in our lab

environment. This discrepancy suggests that our findings may not

fully reflect the intricacies of attentional processes in genuine con-

struction contexts. Additionally, the controlled lab setting, while

advantageous for minimizing external variables and concentrating

on specific stimuli, lacks external factors like background noise,

physical activity, and the stress typical of a working construction

site. These factors are vital in determining the attentional capacity

and fatigue levels of construction professionals. Furthermore, our

study includes a limitation regarding the participant demographic.

The recruited participants, predominantly early-career professio-

nals with managerial roles, such as junior managers, supervisors,

and construction interns, and a gender distribution of 6 males and

14 females, may not fully represent the traditional workforce in the

construction industry. This selection was due to their alignment

with our recruitment criteria and their willingness to participate.

While this contributes to understanding the cognitive impacts

within the evolving construction industry, it also highlights a

gap in representation, particularly of hands-on site workers and

a more traditional gender distribution, which could influence the

generalizability of our results.

Future research could greatly benefit from adopting interdisci-

plinary approaches that incorporate diverse methodologies, such as

eye movement tracking, skin conductance measurements, and vir-

tual reality (VR) simulations. These VR simulations, in particular,

could be designed to more closely mimic the actual conditions of a

construction site, providing valuable insights into how attention

and fatigue are managed in more complex, real-world scenarios.

By integrating a broader range of sensory stimuli and environmen-

tal variables, subsequent studies could achieve a more holistic

understanding of the cognitive processes involved in construction

safety and the fatigue experienced by professionals. Moreover,

addressing the limitations identified in our study, future research

should aim to include a more representative sample of the construc-

tion workforce, encompassing a wider range of job roles and a

gender distribution that mirrors the industry more closely. This ex-

pansion would not only enhance the external validity of the find-

ings but also contribute to a more inclusive understanding of

occupational health and safety in the construction sector.

Conclusion

To address the subjectivity and inconsistency prevalent in research

on attention decline due to fatigue in the construction industry, this

study employed a comprehensive quantitative research methodol-

ogy. Integrating behavioral data, EEG experiments, and subjective

rating scales, we explored the impact of fatigue on attention from a

neuroscientific perspective, with a particular emphasis on effort

regulation. Our analysis of twenty participants performing a 60-min

Oddball cognitive task revealed a slow-fast-slow trajectory in

J. Constr. Eng. Manage., 2024, 150(8): 04024102

attention decline as fatigue intensifies. Key findings include the

identification of proactive strategies in the early stage of fatigue,

where effort effectively counters attention decline. In the middle

stage, as fatigue deepens, resource consumption becomes predomi-

nant, and the efficacy of effort begins to wane. In the late stage,

participants reach an effort threshold, leading to a trade-off between

speed and accuracy.

This study’s innovative contributions lie in elucidating the neu-

ral mechanisms of fatigue-related attention decline in construction

practitioners, including junior managers. We highlight the pivotal

role of effort regulation in maintaining attention and point out the

alternating dominance of resources and effort across different

stages of fatigue. Most importantly, the practical implications for

construction safety are profound: (1) early stage interventions:

Emphasizing the activation of practitioners’awareness and effort

to counter initial attention decline; (2) mid-stage strategies: Imple-

menting scheduled breaks and task rotations to manage deepening

fatigue and resource depletion; and (3) late stage measures: Adopt-

ing motivational measures, complemented by end-of-day reviews,

to mitigate errors and safety risks caused by the decline in attention.

These strategies offer a roadmap for effectively managing fatigue-

induced attention decline at various stages, thereby enhancing

safety management in construction settings. Our findings under-

score the potential of integrating neuroscientific insights into

construction safety practices, providing a foundation for future re-

search and practical applications aimed at reducing accidents and

improving workplace safety.

Data Availability Statement

All data generated or analyzed during the study are available from

the corresponding author by request.

Acknowledgments

This work was supported by the National Natural Science Founda-

tion of China (Grant No. 71972020); the National Social Science

Fund of China (Grant No. 22AZD099); the Graduate Scientific

Research and Innovation Foundation of Chongqing, China

(CYS21036); and the Fundamental Research Funds for the Central

Universities (Grant No. 2020CDJK03ZH04).

References

Akerstedt, T., and M. Gillberg. 1990. “Subjective and objective sleepiness

in the active individual.”Int. J. Neurosci. 52 (Jun): 29–37.

Artis, D. A. Z. 2015. Women as leaders in construction in a northwestern

US city: A multiple-case study of the effectiveness of motivation prac-

tices and success strategies. Minneapolis: Capella Univ.

Boksem, M. A. S., T. F. Meijman, and M. M. Lorist. 2005. “Effects of

mental fatigue on attention: An ERP study.”Cognitive Brain Res.

25 (1): 107–116. https://doi.org/10.1016/j.cogbrainres.2005.04.011.

Bonnefond, A., N. Doignon-Camus, P. Touzalin-Chretien, and A. Dufour.

2010. “Vigilance and intrinsic maintenance of alert state: An ERP

study.”Behav. Brain Res. 211 (2): 185–190. https://doi.org/10.1016/j

.bbr.2010.03.030.

Chen, J., R. Q. Wang, Z. Lin, and X. Guo. 2018. “Measuring the cognitive

loads of construction safety sign designs during selective and sustained

attention.”Saf. Sci. 105 (Feb): 9–21. https://doi.org/10.1016/j.ssci.2018

.01.020.

Csatho, A., D. van der Linden, I. Hernadi, P. Buzas, and G. Kalmar. 2012.

“Effects of mental fatigue on the capacity limits of visual attention.”

J. Cognit. Psychol. 24 (Sep): 511–524.

Delorme, A., and S. Makeig. 2004. “EEGLAB: An open source toolbox for

analysis of single-trial EEG dynamics including independent compo-

nent analysis.”J. Neurosci. Methods 134 (1): 9–21. https://doi.org/10

.1016/j.jneumeth.2003.10.009.

Duncan, C. C., R. J. Barry, J. F. Connolly, C. Fischer, P. T. Michie,

R. Naatanen, J. Polich, I. Reinvang, and C. Van Petten. 2009. “Event-

related potentials in clinical research: Guidelines for eliciting, record-

ing, and quantifying mismatch negativity, P300, and N400.”Clin.

Neurophysiol. 120 (11): 1883–1908. https://doi.org/10.1016/j.clinph

.2009.07.045.

Egeth, H. 1975. “Attention and effort.”Am. J. Psychol. 88 (2): 339–340.

https://doi.org/10.2307/1421603.

Endsley, M. R. 1995. “Toward a theory of situation awareness in dynamic-

systems.”Hum. Factors 37 (1): 32–64. https://doi.org/10.1518/00187

2095779049543.

Faber, L. G., N. M. Maurits, and M. M. Lorist. 2012. “Mental fatigue af-

fects visual selective attention.”PLoS One 7 (10): e48073. https://doi

.org/10.1371/journal.pone.0048073.

Fang, D. P., Z. M. Jiang, M. Z. Zhang, and H. Wang. 2015. “An experi-

mental method to study the effect of fatigue on construction workers’

safety performance.”Saf. Sci. 73 (Apr): 80–91. https://doi.org/10.1016/j

.ssci.2014.11.019.

Fu, H., G. Ye, J. Chen, and P. C. Liao. 2022. “Editorial: Human decision-

making behaviors in engineering and management: A neuropsychologi-

cal perspective.”Front. Neurosci. 16 (Nov): 1062171. https://doi.org/10

.3389/fnins.2022.1062171.

Gergelyfi, M., B. Jacob, E. Olivier, and A. Zenon. 2015. “Dissociation be-

tween mental fatigue and motivational state during prolonged mental

activity.”Front. Behav. Neurosci. 9 (Jul): 176.

Gomes, H., S. Molholm, C. Christodoulou, W. Ritter, and N. Cowan. 2000.

“The development of auditory attention in children.”Front. Biosci.

Landmark 5 (1): D108–D120. https://doi.org/10.2741/Gomes.

Grandjean, E. 1979. “Fatigue in industry.”Occup Environ Med. 36 (Mar):

175–186.

Guo, Z. Z., R. Y. Chen, K. Zhang, Y. R. Pan, and J. H. Wu. 2016. “The

impairing effect of mental fatigue on visual sustained attention under

monotonous multi-object visual attention task in long durations: An

event-related potential based study.”PLoS One 11 (9): e0163360.

https://doi.org/10.1371/journal.pone.0163360.

Hancock, P. A., and J. S. Warm. 1989. “A dynamic-model of stress and

sustained attention.”Hum. Factors 31 (5): 519–537. https://doi.org/10

.1177/001872088903100503.

Hasanzadeh, S., B. Esmaeili, and M. D. Dodd. 2017. “Impact of construc-

tion workers’hazard identification skills on their visual attention.”

J. Constr. Eng. Manage. 143 (10): 04017070. https://doi.org/10.1061

/(ASCE)CO.1943-7862.0001373.

Heinrich, H. W., D. Petersen, and N. Roos. 1950. Industrial accident

prevention: A scientific approach. 2nd ed. New York: McGraw-Hill.

Hockey, G. R. J. 1997. “Compensatory control in the regulation of human

performance under stress and high workload: A cognitive-energetical

framework.”Biol. Psychol. 45 (1–3): 73–93. https://doi.org/10.1016

/S0301-0511(96)05223-4.

Hopstaken, J. F., D. van der Linden, A. B. Bakker, and M. A. J. Kompier.

2015. “The window of my eyes: Task disengagement and mental fatigue

covary with pupil dynamics.”Biol. Psychol. 110 (Jun): 100–106.

https://doi.org/10.1016/j.biopsycho.2015.06.013.

Käthner, I., S. C. Wriessnegger, G. R. Müller-Putz, A. Kübler, and S.

Halder. 2014. “Effects of mental workload and fatigue on the P300,

alpha and theta band power during operation of an ERP (P300) brain-

computer interface.”Biol. Psychol. 102 (Aug): 118–129. https://doi.org

/10.1016/j.biopsycho.2014.07.014.

Ke, J. J., M. Zhang, X. W. Luo, and J. Y. Chen. 2021. “Monitoring

distraction of construction workers caused by noise using a wearable

Electroencephalography (EEG) device.”Autom. Constr. 125 (May):

103598. https://doi.org/10.1016/j.autcon.2021.103598.

Keil, A., S. Debener, G. Gratton, M. Junghöfer, E. S. Kappenman, S. J.

Luck, P. Luu, G. A. Miller, and C. M. Yee. 2014. “Committee report:

Publication guidelines and recommendations for studies using electro-

encephalography and magnetoencephalography.”Psychophysiology

51 (1): 1–21. https://doi.org/10.1111/psyp.12147.

J. Constr. Eng. Manage., 2024, 150(8): 04024102

Khalid, U., A. Sagoo, and M. Benachir. 2021. “Safety management system

(SMS) framework development—Mitigating the critical safety factors

affecting health and safety performance in construction projects.”Saf.

Sci. 143 (Nov): 105402. https://doi.org/10.1016/j.ssci.2021.105402.

Kitani, T. 2011. “Term committee of Japanese society of fatigue science.”

Nihon Hirougakkaishi 6 (Apr): 1.

Koc, K., O. Ekmekcioglu, and A. P. Gurgun. 2023. “Developing a national

data-driven construction safety management framework with interpret-

able fatal accident prediction.”J. Constr. Eng. Manage. 149 (4):

04023010. https://doi.org/10.1061/JCEMD4.COENG-12848.

Kurzban, R., A. Duckworth, J. W. Kable, and J. Myers. 2013. “An oppor-

tunity cost model of subjective effort and task performance.”Behav.

Brain Sci. 36 (6): 661–679. https://doi.org/10.1017/S0140525X

12003196.

Larue, G. S., A. Rakotonirainy, and A. N. Pettitt. 2011. “Driving perfor-

mance impairments due to hypovigilance on monotonous roads.”Accid.

Anal. Prev. 43 (6): 2037–2046. https://doi.org/10.1016/j.aap.2011.05

.023.

Laszlo, S., M. Ruiz-Blondet, N. Khalifian, F. Chu, and Z. P. Jin. 2014. “A

direct comparison of active and passive amplification electrodes in the

same amplifier system.”J. Neurosci. Methods 235 (Dec): 298–307.

https://doi.org/10.1016/j.jneumeth.2014.05.012.

Li, J., H. Li, H. W. Wang, W. Umer, H. Fu, and X. J. Xing. 2019. “Evalu-

ating the impact of mental fatigue on construction equipment operators’

ability to detect hazards using wearable eye-tracking technology.”

Autom. Constr. 105 (Sep): 102835. https://doi.org/10.1016/j.autcon

.2019.102835.

Li, L., L. Zhang, and L. Niu. 2017. “Event-related potentials study of min-

ers’safety attention at different job stages.”J. China Saf. Sci. Technol.

13 (Aug): 96–101.

Lim, J. L., and D. F. Dinges. 2008. “Sleep deprivation and vigilant atten-

tion.”Ann. New York Acad. Sci. 1129 (1): 305–322.

Lorist, M. M. 2008. “Impact of top-down control during mental fatigue.”

Brain Res. 1232 (Sep): 113–123. https://doi.org/10.1016/j.brainres

.2008.07.053.

Luria, A. 1973. The working brain: An introduction to neuropsychology.

New York: Basic Books.

Ma, J., H. Li, X. E. Yu, X. Fang, B. Fang, Z. Y. Zhao, X. C. Huang, S.

Anwer, and X. J. Xing. 2023. “Sweat analysis-based fatigue monitoring

during construction rebar bending tasks.”J. Constr. Eng. Manage.

149 (9): 04023072.

Maynard, S., W. Jones, A. Filtness, A. Gibb, and R. Haslam. 2021.

“Going underground: Fatigue and sleepiness in tunnelling operations.”

Appl. Ergon. 90 (Jan): 103237. https://doi.org/10.1016/j.apergo.2020

.103237.

Mehmood, I., H. Li, Y. Qarout, W. Umer, S. Anwer, H. Wu, M. Hussain,

and M. Fordjour Antwi-Afari. 2023a. “Deep learning-based construc-

tion equipment operators’mental fatigue classification using wearable

EEG sensor data.”Adv. Eng. Inf. 56 (Apr): 101978. https://doi.org/10

.1016/j.aei.2023.101978.

Mehmood, I., H. Li, Y. Qarout, W. Umer, S. Anwer, H. T. Wu, M. Hussain,

and M. F. Antwi-Afari. 2023b. “Deep learning-based construction

equipment operators’mental fatigue classification using wearable

EEG sensor data.”Adv. Eng. Inf. 56 (Apr): 101978. https://doi.org/10

.1016/j.aei.2023.101978.

Naito, Y., H. Ando, and M. Yamaguchi. 2005. “Assessment of traumatic

brain injury patients by WAIS-R, P300, and performance on oddball

task.”Kobe J. Med. Sci. 51 (Aug): 95–105.

Namian, M., F. Taherpour, E. Ghiasvand, and Y. Turkan. 2021. “Insidious

safety threat of fatigue: Investigating construction workers’risk of ac-

cident due to fatigue.”J. Constr. Eng. Manage. 147 (12): 04021162.

https://doi.org/10.1061/(ASCE)CO.1943-7862.0002180.

Nuwer, M. R. 1998. “Fundamentals of evoked potentials and common

clinical applications today.”Electroencephalography Clin. Neuro-

physiol. 106 (2): 142–148. https://doi.org/10.1016/S0013-4694(97)

00117-X.

Reason, J. 1990. Human error. New York: Cambridge University Press.

Sanders, A. F. 1983. “Towards a model of stress and human-performance.”

Acta Psychol. 53 (1): 61–97. https://doi.org/10.1016/0001-6918(83)

90016-1.

Smith, M. R., L. Zeuwts, M. Lenoir, N. Hens, L. M. S. De Jong, and

A. J. Coutts. 2016. “Mental fatigue impairs soccer-specific decision-

making skill.”J. Sports Sci. 34 (14): 1297–1304. https://doi.org/10

.1080/02640414.2016.1156241.

Spronk, M., L. M. Jonkman, and C. Kemner. 2008. “Response inhibition

and attention processing in 5-to 7-year-old children with and without

symptoms of ADHD: An ERP study.”Clin. Neurophysiol. 119 (12):

2738–2752. https://doi.org/10.1016/j.clinph.2008.09.010.

Strüber, D., and J. Polich. 2002. “P300 and slow wave from oddball and

single-stimulus visual tasks: Inter-stimulus interval effects.”Int. J. Psy-

chophysiol. 45 (3): 187–196. https://doi.org/10.1016/S0167-8760(02)

00071-5.

Sun, Y., J. Lim, Z. X. Dai, K. Wong, F. Taya, Y. Chen, J. H. Li, N. Thakor,

and A. Bezerianos. 2017. “The effects of a mid-task break on the

brain connectome in healthy participants: A resting-state functional

MRI study.”Neuroimage 152 (Mar): 19–30. https://doi.org/10.1016/j

.neuroimage.2017.02.084.

Takacs, E., I. Barkaszi, A. Altbacker, I. Czigler, and L. Balazs. 2019.

“Cognitive resilience after prolonged task performance: An ERP inves-

tigation.”Exp. Brain Res. 237 (Dec): 377–388.

Tanaka, M., K. Mizuno, S. Tajima, T. Sasabe, and Y. Watanabe. 2009.

“Central nervous system fatigue alters autonomic nerve activity.”Life

Sci. 84 (7–8): 235–239. https://doi.org/10.1016/j.lfs.2008.12.004.

Techera, U., M. Hallowell, and R. Littlejohn. 2018a. “Worker fatigue in

electrical-transmission and distribution-line construction.”J. Constr.

Eng. Manage. 145 (1): 04018119. https://doi.org/10.1061/(ASCE)CO

.1943-7862.0001580.

Techera, U., M. Hallowell, R. Littlejohn, and S. Rajendran. 2018b.

“Measuring and predicting fatigue in construction: Empirical field

study.”J. Constr. Eng. Manage. 144 (Aug): 04018062.

Umer, W., Y. T. Yu, and M. F. A. Afari. 2021. “Quantifying the effect of

mental stress on physical stress for construction tasks.”J. Constr. Eng.

Manage. 148 (3): 04021204. https://doi.org/10.1061/(ASCE)CO.1943

-7862.0002243.

van der Linden, D., and P. Eling. 2006. “Mental fatigue disturbs local

processing more than global processing.”Psychol. Res. 70 (5): 395–

402. https://doi.org/10.1007/s00426-005-0228-7.

Wang, Y., Y. C. Huang, B. T. Gu, S. H. Cao, and D. P. Fang. 2023. “Iden-

tifying mental fatigue of construction workers using EEG and deep

learning.”Autom. Constr. 151 (Jul): 104887. https://doi.org/10.1016/j

.autcon.2023.104887.

Watford, T. S., W. H. O’Brien, H. R. Koerten, L. M. Bogusch, M. T.

Moeller, R. Sonia Singh, and T. E. Sims. 2020. “The mindful attention

and awareness scale is associated with lower levels of high-frequency

heart rate variability in a laboratory context.”Psychophysiology 57 (3):

e13506. https://doi.org/10.1111/psyp.13506.

Wenger, M. J., and B. S. Gibson. 2004. “Using hazard functions to assess

changes in processing capacity in an attentional cuing paradigm.”

J. Exp. Psychol. Hum. Percept. Perform. 30 (4): 708–719. https://doi

.org/10.1037/0096-1523.30.4.708.

Westbrook, A., and T. S. Braver. 2015. “Cognitive effort: A neuroeconomic

approach.”Cognit. Affect. Behav. Neurosci. 15 (2): 395–415. https://doi

.org/10.3758/s13415-015-0334-y.

Xing, X. J., B. T. Zhong, H. B. Luo, T. Rose, J. Li, and M. F. Antwi-Afari.

2020. “Effects of physical fatigue on the induction of mental fatigue of

construction workers: A pilot study based on a neurophysiological ap-

proach.”Autom. Constr. 120 (Dec): 103381. https://doi.org/10.1016/j

.autcon.2020.103381.

Zhang, Z. T., T. Xiang, H. L. Guo, L. Ma, Z. Y. Guan, and Y. H. Fang. 2023.

“Impact of physical and mental fatigue on construction workers’unsafe

behavior based on physiological measurement.”J. Saf. Res. 85 (Aug):

457–468. https://doi.org/10.1016/j.jsr.2023.04.014.

Zhao, C. L., M. Zhao, J. P. Liu, and C. X. Zheng. 2012. “Electroencepha-

logram and electrocardiograph assessment of mental fatigue in a driving

simulator.”Accid. Anal. Prev. 45 (Feb): 83–90. https://doi.org/10.1016/j

.aap.2011.11.019.

Zijlstra, F. R. H. 1993. “Efficiency in work behavior: A design approach for

modern tools.”Doctoral thesis, Faculty of Philosophy and Technical

Social Sciences, Delft Univ. of Technology.

J. Constr. Eng. Manage., 2024, 150(8): 04024102

ResearchGate has not been able to resolve any citations for this publication.

Deep learning-based construction equipment operators’ mental fatigue classification using wearable EEG sensor data (50 days free access till June 16, 2023: https://authors.elsevier.com/a/1g%7EgP5FA1k5p7e)

Article

Full-text available

Apr 2023
ADV ENG INFORM

Operator attention failure due to mental fatigue during extended equipment operations is a common cause of equipment-related accidents that result in catastrophic injuries and fatalities. As a result, tracking operators' mental fatigue is critical to reducing equipment-related accidents on construction sites. Previously, several strategies aimed at recognizing mental fatigue with adequate accuracy, such as machine learning utilizing EEG-based wearable sensing systems, have been proposed. However, the ability to track operators’ mental fatigue for its implementation on an actual construction site is still an issue. For instance, the mobility and systemic instability of EEG sensors necessitate their application in laboratory settings rather than on actual construction sites. Furthermore, while the machine learning classifiers achieved acceptable accuracy, their input is limited to manually developed EEG features, which may compromise the models’ performance on real construction sites. Accordingly, the current research proposes the viability of a construction site strategy that uses flexible headband-based sensors for acquiring raw EEG data and deep learning networks to recognize operators' mental fatigue. To serve this purpose, a one-hour excavator operation by fifteen operators was conducted on a construction site. The NASA-TLX score was used as the ground truth of mental fatigue, and brain activity patterns were recorded using a wearable EEG sensor. The raw EEG data was then used to develop deep learning-based classification models. Finally, the performance of deep learning models, i.e., long short-term memory, bidirectional LSTM, and one-dimensional convolutional networks, was investigated using accuracy, precision, recall, specificity, and an F1-score. The findings indicate that the Bi-LSTM model outperforms the other deep learning models with a high accuracy of 99.941% and F1-score between 99.917% and 99.993%. These findings demonstrate the feasibility of applying the Bi-LSTM model and contribute to wearable sensor-based mental fatigue recognition and classification, thus enhancing on-site health and safety operations.

Editorial: Human decision-making behaviors in engineering and management: A neuropsychological perspective

Article

Full-text available

Nov 2022

Safety Management System (SMS) framework development – Mitigating the critical safety factors affecting Health and Safety performance in construction projects

Article

Full-text available

Jul 2021
SAFETY SCI

The construction industry is known both for its significance in economic growth and its hazardous nature. The accidents on construction sites not only cause fatalities but also affect project performance severely in term of delayed completion, cost overruns, reduced quality and eventually low productivity. Statistically, poor safety performance is the main cause of the accident on sites due to the number of influencing factors. To improve safety performance, it is inevitable to investigate potential factors involved in safety management. This is a working paper and examines the relative importance of key factors influencing Health and Safety (H&S) performance and the rationale for developing a robust safety management system (SMS) that migrates all factors into one framework. This paper adopts an empirical research methodology based on literature review and secondary data gathered systematically from peer-reviewed journals. There are around sixty H&S factors and these have been assigned to cluster leadings forming six groups namely: ‘organisational’, ‘managerial’, ‘legislative’, ‘social’, ‘environmental’ and ‘personnel’ factors. In developing the rationale for the safety management system (SMS) framework it has become apparent that the effective safety performance can only be achieved through effective (1) implementation of safety regulations, (2) leadership, (3) safety planning, (4) safety compliance, (5) performance measurement, (6) risk assessment, (7) safety inspection, and (8) Safety Culture. These factors are interrelated with each other and they cannot be isolated, however, in order to significantly improve the safety performance target on construction projects, there is a need to re-alignment and re-balance the priorities assigned to factors influencing safety performance.

Sweat Analysis-Based Fatigue Monitoring during Construction Rebar Bending Tasks

Article

Sep 2023

Impact of physical and mental fatigue on construction workers’ unsafe behavior based on physiological measurement

Article

May 2023
J SAFETY RES

Introduction: Construction worker fatigue is an important factor leading to unsafe behavior, a major cause of construction accidents. Uncovering the impact mechanism of fatigue on workers' unsafe behavior can prevent construction accidents. However, it is difficult to effectively measure workers' fatigue onsite and analyze the impact of worker fatigue on their unsafe behavior. Method: This research analyzes the relationship between the physical and mental fatigue of construction workers and their unsafe behavior via physiological measurement based on a simulated experiment on handling tasks. Results: It is found that: (a) both physical fatigue and mental fatigue have negative effects on workers' cognitive ability and motion ability, and the negative effects are more serious under the combination of the two types of fatigue; (b) mental fatigue can easily change workers' risk propensity, making them more willing to face risks, and in a state of the two types of fatigue, they are more likely to make choices with less pay and higher risk; (c) the number of signal identification errors is positively correlated with LF (low frequency)/HF (high frequency), and negatively correlated with the standard deviation of normal-to-normal intervals (SDNN), while the number of footstep control errors is negatively correlated with the time elapsed between two successive R waves (RR interval) and skin temperature (SKT). Practical applications: These findings can enrich construction safety management theory from a perspective of quantified fatigue and facilitate safety management practices on construction sites, thus contributing to the body of knowledge and practices of construction safety management.

Identifying mental fatigue of construction workers using EEG and deep learning

Article

Apr 2023
AUTOMAT CONSTR

Construction workers frequently experience mental fatigue owing to the high cognitive load of their tasks in a dynamic, complex environment, diminishing their cognitive ability and mobility and necessitating monitoring to ensure safety. Traditional fatigue evaluations rely on the subjective judgment of site managers and workers; this study developed a wireless device to objectively monitor the mental fatigue of construction workers using electroencephalogram (EEG) signals. The EEG signals of 16 construction workers were recorded while they performed a continuous physical–cognitive task, and the resulting EEG time–frequency–energy data were processed by a continuous wavelet transform and convolutional neural network to identify mental fatigue states without requiring manual feature extraction. The cognitive fatigue state classifications provided by the proposed framework were shown to match the self-reported fatigue states with 88.85% accuracy. This method can therefore facilitate real-time, continuously updating fatigue identification to support safety management on construction sites.

Developing a National Data-Driven Construction Safety Management Framework with Interpretable Fatal Accident Prediction

Article

Feb 2023

Occupational accidents are frequent in the construction industry, containing significant risks in the working environment. Therefore , early designation, taking preventive actions, and developing a proactive safety risk management plan are of paramount significance in managing safety issues in the construction industry. This study aims to develop a national data-driven safety management framework based on accident outcome prediction, which helps anatomize precursors of fatalities and thereby minimizing fatal accidents on construction sites. A national data set comprising 338,173 occupational accidents recorded in the construction industry across Turkey was used to develop a data-driven model. The random forest algorithm coupled with particle swarm optimization was used for the prediction and the interpretability of the proposed model was augmented through the game theory-based Shapley additive explanations (SHAP) approach. The findings showed that the proposed algorithm achieved satisfactory model performances for detecting construction workers who might face a fatality risk. The SHAP analysis results indicated that both company (such as number of past accidents and workers in the company) and worker-related (such as age, daily wage, experience, shift, and past accident of the workers) attributes were influential in identifying fatalities by detecting which workers might face fatal accidents under which conditions. A construction safety management plan was developed based on the analysis results, which can be used on construction sites to detect workers/conditions that are most susceptible to fatalities. The findings of the present research are expected to contribute to orchestrating effective safety management practices in construction sites by characterizing the root causes of severe accidents.

Insidious Safety Threat of Fatigue: Investigating Construction Workers’ Risk of Accident Due to Fatigue

Article

Dec 2021

Monitoring distraction of construction workers caused by noise using a wearable Electroencephalography (EEG) device

Article

May 2021
AUTOMAT CONSTR

In the construction environment with high attention requirements, distraction is the main cause of unsafe behavior and safety performance degradation. However, few studies have focused on distraction's cognitive features and how to monitor it objectively in the construction workplace. To fill the research gap, the present study examined the correlation between distraction and brain activity using an Electroencephalography (EEG) device, intending to provide an approach for objectively monitoring worker distraction. In the simulated hazards identification activity, sustained attention to response task and dual-task paradigms have been employed to induce distraction combined with noise interference. Twenty-seven subjects participated in the experiment to identify whether a hazardous opening exists or not in the workplace in the shown images. The EEG waves were recorded and divided into two groups according to task performance: focused and distracted. Through feature calculation and extraction, it was found that beta and gamma powers in the left temporal and right pre-frontal cortex can distinguish these two statuses, particularly in channels T7 and AF4. The indicators can be considered as an objective evaluation of an individual's sustained attention and attention failures. The developed indicators located in specified brain zones can also be used as a reference for attention training. By providing safety managers with attention status about the workers in high-risk workplaces, distraction detection contributes to control and regulate work error and improper operation, which can extend to apply in other attentive jobs like drivers, pilots, surgeons, and lifeguards.

Effects of physical fatigue on the induction of mental fatigue of construction workers: A pilot study based on a neurophysiological approach

Article

Dec 2020
AUTOMAT CONSTR

Within a dynamic and complex working environment, fatigue statuses (involving physical and mental fatigue) of workers on construction sites tend to have a more serious impact on work performance than general workplaces. To improve safety management on sites, valid fatigue management measures for workers are urgently required. Specifically, there are construction activities requiring both physical and cognitive effort. As a critical premise for putting forward feasible fatigue management measures, correlations between physical and mental fatigue on work performance should be identified. This research explored the effects of physical fatigue on the induction of mental fatigue of construction workers, by adopting a pilot experimental method. Manual handling tasks of different intensities were firstly designed for stimulating certain expected physical fatigue statuses. A cognition-required risk identification task was then arranged for inducing mental fatigue, during which a wearable electroencephalogram (EEG) sensor was utilized for fatigue detection and measurement. Through a comprehensive data analysis method based on EEG rhythms, it was found that the high physical fatigue can significantly accelerate the induction of mental fatigue. Considering the resource allotment, more vigilant and attentional resources were required during the intensive manual handling tasks for the highly controlled limbs and the mind to steps. Thus, additional resources were invested to maintain the same level of cognitive performance in the risk identification tasks, which led to the increased mental fatigue. In practice, the heavy physical task can be regarded as one of the factors affecting the development of mental fatigue status, and therefore impairing cognitive functioning and other mental performances of the brain. The pilot study results provided a reference for fatigue management of construction workers to promote comprehensive safety management on construction sites.

Impact of Cognitive Fatigue on Attention and the Implications for Construction Safety: A Neuroscientific Perspective

Abstract and Figures

Recommended publications

Impact of Cognitive Fatigue on Attention and the Implications for Construction Safety: A Neuroscient...

Exploring the Internal Influence Mechanism of Group Safety Behavior of Construction Workers: Qualita...

Sleep Duration and Results of the Psychomotor Vigilance Test in Construction Workers: A Preliminary...

Insights into the mechanism of construction workers' unsafe behaviors from an individual perspective