Online Hazard Recognition Training: Comparative Case Study of Static Images, Cinemagraphs, and Videos

Abstract

This research studied the use of cinemagraphs for online training purposes to enhance the hazard recognition of construction “struck-by” hazards. Cinemagraphs (i.e., partially animated videos that loop over time) have been identified as an effective method to direct attention in video content, making it potentially beneficial in remote construction safety training applications where the instructor has no control over what the trainee is observing. To explore the effectiveness of different methods to direct attention in video content for hazard-recognition training purposes, three degrees of animations in narrated videos were compared: (1) static images: regular fixed pictures; (2) partially animated images: cinemagraphs; and (3) fully animated images: traditional videos. A case study was utilized for training students to recognize a select set of struck-by hazards, employing a between-subjects experimental design. The case study contained six potentially fatal struck-by scenarios, independently demonstrated through videos to each group of participants under each experimental condition. Posteriorly to the training, the subjects were evaluated using an online assessment test for stuck-by hazard identification. Through this experimental methodology, measures regarding the hazard identification index (HII), attitudes, cognitive absorption, and transportability were collected from 145 study participants. The results of this investigation did not detect statistical differences in the average hazard identification index scores using narrated static images (average=67.1%; standard deviation (SD) = 19%), cinemagraphs (average=61.5%; SD = 20%), or videos (average=64.8%; SD = 15%). Moreover, the survey responses also suggested that participants across all the experimental conditions had equally positive attitudes, engagement, and sense of being transported to the scenario location. Due to the lack of differences detected in this study across experimental conditions, it was concluded that cinemagraphs provide online training content creators another viable technique with equivalent benefits to traditional methods.