Why safety assurances lead to higher risk perceptions: A conversational approach
AltmetricsView Usage Statistics
Safety assurances are meant to decrease people's perceptions of risk. This paper explored the notion that safety assurances may produce the opposite effect - higher perceptions of risk regarding the object of assurance. The project pursued two objectives: 1) to provide a theoretical analysis of the cognitive mechanisms facilitating the unintended effect of safety assurances, and 2) to obtain empirical evidence for this effect in the context of risk communication. Pragmatic theories of speech acts, conversational logic, and common ground provided a theoretical framework that explains this paradoxical effect as the outcome of a hearer's inferences about the meaning of a speaker's communication. Based on a review of empirical evidence for the operation of conversational rules in institutional settings, the paper explored contextual factors that would contribute to the unintended effects of safety assurances in risk communication. Using a simulation approach, a series of four experiments examined the effect of safety assurances on the perceptions of risk regarding future events in a local community. Studies 1 (N = 141) and 3 (N = 411) showed that participants who read an announcement of an upcoming transport of mining wastes through the area, complemented by a safety assurance, perceived the campaign to be more harmful than the participants in control condition who read the announcement without assurance. In Study 4 (N = 516), a similar pattern of higher perceptions of risk was observed for participants who read an article about a restaurant, in which the owners assured the safety of their food. In Study 2 (N = 153), participants read assurances about safety of their drinking water but no significant differences in perceived risk were observed. The studies examined a number of variables that had been found to moderate perceptions of risks. Consistent with prior research, compared to men, women perceived significantly higher risks on several dimensions. The length of residency in the local community moderated the extent of the safety assurances effect. Consistent with the prediction of the conversational framework, long-term residents inferred higher risks and less benefit from future events in response to safety assurances compared to short-term residents. Studies 3 and 4 explored the conditions under which safety assurances would result in lower perceptions of risks. The discussion focuses on the implications of the findings for conversational framework and risk communication and elaborates on methodological limitations as well as directions for future research.