The uneven power of pictures: How individual perceived risk and uncertainty are influenced by photographs, non-photographs and the language used to describe them

Ansel Adams, a famous American photographer and dedicated environmentalist, captured the power of photographs, “Not everybody trusts paintings but people believe photographs” (Zeegen, 2010, p.102). Indeed, photographs are a powerful visual tool that provides evidence that an object existed in one brief moment in time (Messaris, 1994; Sontag, 1977). This visual evidence is particularly useful in a risk communication context providing visual proof that a risk, such as a tornado, exists, and that action is imminent. With photo capturing technologies, such as digital cameras, camera phones, and more recently, Google glass, photographs of an event are often available and accessible via social media tools. Due to their believable characteristics, during risk events photographs are prominently displayed during live television broadcasts. These visuals provide useful, certain information during an unfolding crisis that a non-photograph may not be able to provide. The concern is that even with all available technology photographs are not always possible to attain nor does certainty exist during all events. The risk itself may be invisible, such as a high ozone day causing air quality concerns, or the risk may be present, but not visible at that moment, for example, a rain wrapped tornado. Scientists must use non-photographic images to relay and visually capture these types of risks. When visual confirmation of a tornado is not available, for example, meteorologists must use radar to predict where and when one could form. But, what are the effects of these different types of visuals on audience perception? And further, how does a scientists' language impact the effectiveness of these visuals? Using real-time response measurement, this research examines individuals perceived uncertainty and risk of information presented during tornado news coverage. Specifically, this study explores whether individuals perceive more or less uncertainty and risk based on the visual information presented in the videos. Further, the videos were transcribed, and the language was coded for themes of threat, impact and behavioral messages. Combining the coding with the visual types reveals interesting perceptual patterns. Photographs in all conditions of language prompted higher perceived risk and lower perceived uncertainty (more certainty) than the non-photographic images. Within non-photographic images, there were no significant differences between threat, impact and behavioral messages. Further results, explanations of limitations and next steps will be discussed.