David Kellen, Clintin Davis-Stober, and I have a new paper out on risk attitudes (see below for title, abstract, and full reference).

In this work, we help further muddling the field by showing that individual and age differences are not consistent across different types of decision problems, such as those between two risky options involving gains, or losses, or both (as opposed to a risky option vs. a sure outcome which has been the focus of most past work). For example, we show that in some but not all gamble problems, age differences can be found…

Individual classification of strong risk attitudes: An application across lottery types and age groups

Empirical evaluations of risk attitudes often rely on a weak definition of risk that concerns preferences towards risky and riskless options (e.g., a lottery vs. a sure outcome). A large body of work has shown that individuals tend to be weak risk averse in choice contexts involving risky and riskless gains but weak risk seeking in contexts involving losses, a phenomenon known as the reflection effect. Recent attempts to evaluate age differences in risk attitudes have relied on this weak definition, testing whether the reflection effect increases or diminishes as we grow older. The present work argues that weak risk attitudes have limited generalizability and proposes the use of a strong definition of risk that is concerned with preferences towards options with the same expected value but different degrees of risk (i.e., outcome variance). A reanalysis of previously-published data and the results from a new study show that only a minority of individuals manifests the reflection effect under a strong definition of risk, and that, when facing certain lottery-pair types, older adults appear to be more risk seeking than younger adults.

Kellen, D., Mata, R., & Davis-Stober, C.P. (2017). Individual classification of strong risk attitudes: An application across lottery types and age groups. Psychological Bulletin & Review. doi: 10.3758/s13423-016-1212-5.