Development and testing of a personalized hazard-recognition training intervention

Unrecognized or unmanaged hazards can expose workers to unanticipated safety risk and can potentially result in catastrophic safety incidents. Unfortunately, recent research has demonstrated that a large proportion of safety hazards remain unrecognized in construction workplaces. To improve hazard-recognition levels, employers adopt a variety of safety and hazard-recognition training programs. However, desirable levels of hazard recognition have not been achieved, and the expected benefits from training have not been attained. Such failure in training efforts have generally been attributed to the adoption of poor and ineffective training practices. While efforts are being undertaken to address these issues, construction research has not focused on developing or evaluating personalized training solutions that are customized to the learning needs of individual workers. To advance theory and practice, the objective of this study was to develop the first personalized training strategy targeted at improving hazard-recognition levels. The objective was accomplished by a collaborative effort involving two industry experts and three academic researchers, along with guidance from training literature. The training strategy incorporates important elements known to improve stimuli or threat detection in domains including medicine, the military, and aviation. The elements include (1) visual cues to aid systematic hazard search, (2) personalized hazard-recognition performance feedback, (3) personalized eye-tracking visual attention feedback, and (4) metacognitive prompts that trigger the adoption of remedial measures. After development, the effectiveness of the training strategy in improving hazard recognition was empirically evaluated using the nonconcurrent multiple-baseline testing approach. The findings of the study showed that the participating workers on average were able to identify only 42% of hazards prior to the introduction of the intervention; but were able to recognize 77% of hazards in the intervention phase. The findings of this study will be of interest to practicing professionals seeking to improve hazard-recognition levels within construction.