Patterns, Thinking, and Cognition: A Theory of Judgment

Abstract

For decades, both policymakers and analysts have been frustrated by sharp and stubborn conflicts between expert and lay perceptions on issues of environmental risk. For example, most experts - even those opposed to nuclear power on other grounds - would see precautions like those now in place as adequate to protect against risks from nuclear waste. But the public finds that very hard to believe. Similar sharp conflicts of expert/lay intuition are evident on a wide range of risk issues, from the safety of bendictin as a treatment for morning sickness to the safety of irradiation of food to destroy microorganisms. In Dealing with Risk, Howard Margolis explores the expert/lay rift surrounding such contentious issues and provides a provocative new account. The usual explanation of expert/lay conflicts is that experts are focused only on a narrow notion of risk - such as potential fatalities - but lay intuition is concerned about a wide range of further concerns, such as fairness and voluntariness of exposure. Margolis argues that this rival rationalities view in a fundamental way misses the point of these controversies, since the additional dimensions of lay concern often are more plausibly interpreted as reflections of lay concern than as causes. Margolis argues that risk assessment typically involves weighing a broad range of often complicated trade-offs between costs and benefits. As laypersons, however, we are by definition forced to make judgments on complex matters beyond the scope of our normal experience. Especially in cases involving potential danger, we frequently discount nuance and respond more viscerally. Cognitively we fall back on default responses, all-purpose intuitionssuch as better safe than sorry or ventured, nothing gained. Such intuitions don't admit of careful balancing of pros and cons, and lay opinion consequently becomes polarized and at odds with the expert view.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 161 citations exist upstream.