Avoiding Older Construction Workers' Unsafe Behaviors Related to Self-Cognitive Bias

Abstract

The construction industry is a crucial pillar of the global economy, yet it is also a high-risk sector with frequent safety accidents. With the continuous expansion of the global construction market and the aging of the construction workforce, the occupational safety of older construction workers has become an issue that cannot be ignored. In fact, unsafe behaviors of construction workers are often the main cause of safety accidents. Consequently, the occupational safety of older construction workers and the safety behaviors of construction workers have gradually become hot topics of global scholarly attention. Existing research has identified some factors influencing the occupational safety of older construction workers and explained the mechanisms behind unsafe behaviors from various perspectives. However, there has been a lack of focus on the impact of self-cognitive biases of older construction workers on their unsafe behaviors. Self-cognitive bias refers to the systematic bias in individuals' assessment of their abilities. Due to long-term accumulated work experience, habitual thinking patterns, and a lack of timely updates on new technologies and safety standards, older construction workers often exhibit self-cognitive biases, which are closely related to unsafe behaviors. However, existing research lacks an analysis of the connotation and dimensional structure of self-cognitive biases of older construction workers and an explanation of their impact mechanism on unsafe behaviors. Based on this, this study aims to clarify the connotation and dimensional structure of self-cognitive biases of older construction workers, establish an objective measurement method for these biases, analyze their influencing factors, reveal the pathways through which they affect workers' unsafe behaviors, simulate the evolutionary patterns of self-cognitive biases and unsafe behaviors of older construction workers under different management strategies, and provide management insights.

Firstly, the study uses qualitative analysis methods to explain the connotation and dimensional structure of self-cognitive biases of older construction workers and constructs a conceptual model of their impact mechanism on unsafe behaviors. Then, based on event-related potential experiments and the semantic ambiguity paradigm, it is proven that older construction workers generally exhibit self-cognitive biases. An objective measurement method for these biases based on N400 amplitude is established, and factors such as age, safety training, and health check habits affecting self-cognitive biases are explored. Subsequently, referring to existing mature scales and interview records, measurement scales for various variables in the conceptual model of the impact mechanism of self-cognitive biases of older construction workers on their unsafe behaviors are designed and perfected, and the reliability and validity of the scales are tested. Based on observational data from questionnaire surveys, the study uses statistical methods such as structural equation modeling and regression analysis to validate the established conceptual model, revealing the mechanism by which self-cognitive biases of older construction workers affect their unsafe behaviors. Finally, based on the results of model testing, a system dynamics model is constructed to simulate the dynamic evolution of self-cognitive biases and unsafe behaviors of older construction workers under different management strategies. The study finds: (1) Self-cognitive biases of older construction workers are a multi-dimensional complex concept, which can be divided into three dimensions: underestimation of physical function degradation, underestimation of cognitive function degradation, and overestimation of the role of experience; (2) Older construction workers generally exhibit self-cognitive biases in these three dimensions, which are affected by factors such as age, work-related injury experience, health check habits, and safety training. Older individuals have smaller self-cognitive biases in physical and cognitive functions than younger ones. Those with work-related injury experience have smaller self-cognitive biases in physical and cognitive functions and the role of experience than those without. Those with health check habits have smaller self-cognitive biases in physical function than those without. Those with safety training experience have smaller self-cognitive biases in the role of experience than those without; (3) Self-cognitive biases of older construction workers are related to unsafe behaviors. Underestimation of physical function degradation affect workers' unsafe behaviors through safety attitudes, self-efficacy, and risk perception. Underestimation of cognitive function degradation affect workers' unsafe behaviors through self-efficacy and risk perception. Overestimation of the role of experience affect workers' unsafe behaviors through safety attitudes, self-efficacy, safety knowledge, and risk perception. Management practices can play a significant moderating role in the direct or indirect impact of self-cognitive biases of older construction workers on their unsafe behaviors; (4) System dynamics simulation confirms that providing regular health checks, enhancing safety training, and adjusting work arrangements can weaken the self-cognitive biases of older construction workers and reduce their unsafe behaviors. The effect of combined strategies is better than that of individual strategies. The research results expand the cognitive bias theory, deepen the research on the causation mechanism of unsafe behaviors of older construction workers, and provide theoretical basis and decision-making reference for construction safety management.

The innovation of the thesis lies in the comprehensive analysis of the multi-dimensional structure of self-cognitive biases of older construction workers, the introduction and application of neuroscience tools, the in-depth revelation of the multi-level mechanism of the relationship between self-cognitive biases and unsafe behaviors of older construction workers, and the application of system dynamics models as policy simulators to simulate the dynamic evolution of self-cognitive biases and unsafe behaviors of older construction workers. Future research should consider cross-cultural and regional comparative studies, long-term tracking studies and dynamic analysis, and innovative applications of technologies and tools such as deep learning and virtual reality.

Date of Award	4 Sept 2025
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Jianling HUANG (External Supervisor) & Xiaowei LUO (Supervisor)