Academic journal article Memory & Cognition

Risk Avoidance: Graphs versus Numbers

Academic journal article Memory & Cognition

Risk Avoidance: Graphs versus Numbers

Article excerpt

There have long been speculations that graphical and numerical presentations of risk statistics differ in their impact on people's willingness to pursue actions that could harm or even kill them. But research has been unclear about the processes whereby the pictorial character of graphical displays per se might affect those risky decisions or even whether such effects actually occur. In two studies, we demonstrate that the pictorial nature of a graphical risk display can, indeed, increase risk avoidance. This increase is associated with a heightened impression of the riskiness of less safe alternatives. The results suggest that this picture-driven, intensified sense of riskiness, in turn, rests on two kinds of mechanisms: one cognitive, the other affective. Cognitively, pictorial presentations impose weaker upper bounds on people's internal representations of the chances that riskier alternatives will bring about actual harm. Affectively, pictures ignite stronger, more aversive negative associations with riskier options and their outcomes.

Since the 1940s (e.g., Preston & Baratta, 1948), seemingly countless studies have sought to unravel the processes by which people choose among alternatives that differ in their riskiness. And that work has yielded considerable insights (see, e.g., Hastie, 2001). In many laboratory decision-making studies, to this day, the chances associated with the various outcomes that might result from risky alternatives are presented to participants numerically-for example, as probabilities that particular amounts of money will be gained or lost. In real life, such chances are, indeed, sometimes displayed to deciders that way, but sometimes they are represented differently-for instance, graphically. In a floor debate about motorcycle helmet laws, a legislator could show the numbers of helmeted and helmetless cyclists reduced to vegetative states in highway accidents over the past 5 years. Alternatively, the legislator could show a graph depicting the same statistics. Which display would be more persuasive in getting the legislator's colleagues to vote for stricter helmet requirements (the safer alternative)? More important from a theoretical perspective, how could we explain how logically equivalent pictorial and numerical descriptions might differentially affect choices among risky alternatives? Put another way, what would such effects imply about the nature of the fundamental processes responsible for risk avoidance behavior?

One definition of risk is that it is the possibility of loss (Yates & Stone, 1992, p. 4). Risk increases both with increases in the chances of potential losses and with increases in their severity. Thus, riding a motorcycle without a helmet would be riskier than otherwise if the probability of injury were greater or if the resulting injuries were more serious. There have long been speculations that graphs are especially effective tools for risk communication (e.g., Covello, Sandman, & Slovic, 1988; Keeney & von Winterfeldt, 1986). If true, this would imply that decisions predicated on graphical displays should be different from those based on other characterizations. Indeed, the expectation has been that the decisions should be better, although it is by no means obvious what better ought to mean.

Surprisingly, there was no empirical evidence bearing on speculations such as these before the experiments of Stone, Yates, and Parker ( 1997). In those experiments, the chances of loss (i.e., injuries) associated with various alternatives were displayed either numerically or graphically, in terms of stick figures, asterisks, bar graphs, or faces. The results indicated that graphical risk displays induce significantly greater risk avoidance than do numerical displays. Although Stone et al.'s (1997) experiments demonstrated the potential impact of graphs on risk avoidance, it remained a mystery precisely how such effects come about.

Then, on the basis of several experiments, Stone et al. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.