Both analysis and feelings? The influence of risk beliefs on holistic risk judgments through dual systems using the ESSA model

Abstract

Considerable previous research has examined factors that make up risk perception (its content), and the mechanisms through which risk is perceived (the process(es)). Content-based models have the advantage of identifying specific beliefs that may be out of line with expert judgments, while process-based models are better able to isolate specific mechanisms through which beliefs operate. Despite the advantages of both types, to date most models of risk perception have focused on either one or the other. Here we present and test a revised version of the content-based model tested in Walpole and Wilson (2021) that includes exposure to hazards, susceptibility to consequences and the magnitude of consequences (risk beliefs) and affective responses to hazards. We arrange these concepts in a mediated format that is capable of representing the deliberative and affective processes through which risk beliefs are integrated into holistic risk judgments. We test two versions of the model using a sample of undergraduate students (n = 422), one in which risk beliefs operate independently of one another, and another where the effects of risk beliefs interact with one another. We find evidence to suggest that affective responses are a plausible mediator between risk beliefs and holistic risk judgments, suggesting that this arrangement of the model is capable of representing both deliberative and affective pathways for apprehending risk. Further we find evidence to suggest that the effects of risk beliefs augment one another in elevating holistic risk judgments through both pathways. We close by discussing implications for risk communication and directions for future research.

Keywords:

• Risk perception
• dual-systems
• affect

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 We chose to focus on personal risk rather than general risk because it is likely to be the more relevant of the two for developing a model suited to predicting individual protective actions. This is due to the substantial number of factors that could interrupt the translation of general risk perceptions into personal protective action (see for instance, Lindell and Perry 2012; Wachinger et al. 2013).

2 Following these analyses, we also conducted a conditional process analyses examining moderation of the indirect effects of risk beliefs. Because the results are broadly similar to those reported in the first two sets of analyses, they have been omitted for brevity. Results of these analyses are available from the corresponding author on request.

3 Because susceptibility lies in the ‘middle’ of the causal chain of hazard beliefs (i.e. susceptibility is conditional on exposure and severity is conditioned on susceptibility), it seemed a logical choice to present as the ‘focal predictor’ in the tables and figures presented below. These tables and figures most readily portray the impacts of susceptibility and how those effects are altered by differing levels of the other two factors.