Assessment of Warning Comprehensibility Utilizing Subjective and Physiological Measures

Abstract

Accidents are multi-causal and can be caused by one or more factors such as human errors, machine failures, inadequate safety information, etc. Despite the fact that human errors cannot be completely prevented, various attempts had been made to minimize the likeliness of human errors, one of which is the utilization of safety signs. Consequently, it is important for safety signs to be well designed and comprehended correctly for them to be able to convey warning messages to their prospective users. Faulty safety sign design could lead to increased potential risks in public places.

This thesis was aimed to propose a comprehensive approach in safety sign evaluation by integrating physiological measures and nontraditional subjective measures that have not been used in previous safety sign evaluation studies, in addition to the existing subjective evaluation methods such as guessability score and cognitive sign features. The secondary objective of this thesis was to investigate the effect of cultural differences in safety sign comprehension. In addition, this study also investigated the utilization of users’ feedback and ergonomic design principles in safety sign redesign attempt. In total, 100 participants (forty Westerners and sixty Koreans) participated in the first major experiment which was focused on investigating cultural difference effects during safety sign comprehension. A set of newly introduced water-sport safety signs was used in this study. In addition, a redesign attempt was done on the poorly comprehended signs by utilizing participants’ feedback from both groups, while taking into account ergonomic principles of good design. The second major experiment was aimed to fill the existing gaps in traditional safety sign evaluating methods, by investigating physiological measures (heart rate, eye tracking measures, EEG signal) and nontraditional subjective measurements (perceived cognitive load and compliance to ergonomic design principles) which were hypothesized to reflect cognitive load elicited by a safety sign. A total of forty two Koreans participated in this study, and two sets of frequently used safety signs (traffic signs and healthcare symbols) were used as stimuli.

The first experiment results suggested that people who come from different cultural origins may have different perception and understanding regarding a set of safety signs. Therefore, findings from our first study justified the importance of designing multi-culturally compatible safety signs. Considering ergonomic design principles as well as participants’ feedback was also proven to be noteworthy in safety sign redesign attempt. The second experiment provided evidences that several physiological measures such as eye tracking measures and EEG spectral power in several brain regions were able to reflect a safety sign comprehensibility level in the form of cognitive load. Some nontraditional subjective measures investigated in the second experiment such as perceived cognitive load and compliance to ergonomic design principles were also able to reflect sign comprehensibility. Among the physiological measures and the nontraditional subjective measures investigated in the second major experiment, several measures such as: blinking rates, maximum and average pupil diameter, time to first fixation, EEG spectral power in prefrontal and visual cortex, perceived cognitive load, as well as compliance to ergonomic design principles were found to be consistent when applied to different types of safety signs. Consequently, we assumed that these measures could be applied across different types of safety signs. As a follow on to our study’s findings, we proposed a framework of a comprehensive safety sign evaluation utilizing the consistent measures investigated in the second major experiment, in addition to using the traditional subjective evaluation methods. We also included other factors (e.g., cultural difference, cognitive features, and ergonomic design principles) which were proven to play an important role in a safety sign redesign attempt, in our proposed framework. By this, we hope to provide adequate amount of evidences concerning a particular sign’s comprehensibility level and fill the existing gaps in the traditional safety sign evaluation and redesign approach.