On the Relationship between Cognitive Ability and Risk Preference.

Many decisions of individuals involve a combination of internal preferences and mental processes related to cognitive ability. As Frederick (2005) argued in this journal, "there is no good reason for ignoring the possibility that general intelligence or various more specific cognitive abilities are important causal determinants of decision making." Since then, a number of empirical studies have focused on the relationship between cognitive ability and decision-making in different contexts. This paper will focus on the relationship between cognitive ability and decision-making under risk and uncertainty. Taken as a whole, this research indicates that cognitive ability is associated with risk-taking behavior in various contexts and life domains, including incentivized choices between lotteries in controlled environments, behavior in nonexperimental settings, and self-reported tendency to take risks. We begin by clarifying some important distinctions between concepts and measurement of risk preference and cognitive ability. In particular, complexity and possible confusions arise because observed measures of risk preference and cognitive ability are used to represent the latent characteristics of these concepts. We discuss the substantial (and somewhat implausible) range of assumptions that need to be satisfied in order to be able to interpret a correlation between measures of risk preference and cognitive ability as a relationship between latent risk preference and latent cognitive ability. Drawing causal inferences from such relationships raises additional challenges. We go on to argue that it is nevertheless important and valuable to study whether cognitive ability is related to measured risk preference (see also Dohmen, Falk, Huffman, and Sunde 2010). Risk preference is typically measured by risky behavior (actual or self-reported). If risky behavior varies systematically with cognitive ability, this may reinforce or counteract the impact of cognitive ability on life outcomes, depending on the nature of the correlation. If there is a relationship, it also becomes important to control for cognitive ability when relating life outcomes to standard revealed preference measures of risk preference. If cognitive ability has a causal impact on measured risk preference, it is important to understand the mechanism, and some intriguing policy implications arise. We then take stock of what is known empirically on the connections between cognitive ability and measured risk preferences, looking at studies using real-world risky behavior, experimental measures of risky choice, and self-reported measures of willingness to take risks. One pattern that emerges frequently in these studies is that cognitive ability tends to be positively correlated with avoidance of harmful risky situations, but it tends to be negatively correlated with risk aversion in advantageous situations. This suggests that the relationship between cognitive ability and risk taking has a reinforcing effect on economic outcomes. There is also intriguing emerging evidence that measured risk preference is particularly strongly related to certain facets of cognitive ability, those that facilitate quantitative problem solving, with implications for understanding mechanisms and possibly for better targeting policy interventions. We conclude by discussing perspectives for future research, in particular the scope for the development of richer sets of elicitation instruments and measurement across a wider range of concepts. We also consider progress in neuroscience, but conclude that at present that field still seems relatively far from allowing definitive conclusions about latent risk preference and cognitive ability. Nevertheless, the existing empirical evidence suggests that interventions to influence cognitive ability, should they be possible, might have spillovers on risky choice.

How people know their risk preference.

People differ in their willingness to take risks. Recent work found that revealed preference tasks (e.g., laboratory lotteries)-a dominant class of measures-are outperformed by survey-based stated preferences, which are more stable and predict real-world risk taking across different domains. How can stated preferences, often criticised as inconsequential "cheap talk," be more valid and predictive than controlled, incentivized lotteries? In our multimethod study, over 3,000 respondents from population samples answered a single widely used and predictive risk-preference question. Respondents then explained the reasoning behind their answer. They tended to recount diagnostic behaviours and experiences, focusing on voluntary, consequential acts and experiences from which they seemed to infer their risk preference. We found that third-party readers of respondents' brief memories and explanations reached similar inferences about respondents' preferences, indicating the intersubjective validity of this information. Our results help unpack the self perception behind stated risk preferences that permits people to draw upon their own understanding of what constitutes diagnostic behaviours and experiences, as revealed in high-stakes situations in the real world.

No evidence for an effect of testosterone administration on delay discounting in male university students.

Intertemporal choices between a smaller sooner and a larger delayed reward are one of the most important types of decisions humans face in their everyday life. The degree to which individuals discount delayed rewards correlates with impulsiveness. Steep delay discounting has been associated with negative outcomes over a wide range of behaviors such as addiction. However, little is known about the biological foundations of delay discounting. Here, we examine a potential causal link between delay discounting and testosterone, a hormone which has been associated with other types of impulsive behavior. In our double-blind placebo-controlled study 91 healthy young men either received a topical gel containing 50 mg of testosterone (N=46) or a placebo (N=45) before participating in a delay discounting task with real incentives. Our main finding is that a single dose administration of testosterone did not lead to significant differences in discount rates between the placebo and the testosterone group. Within groups and in the pooled sample, no significant relationship between testosterone and discount rates was observed. At the same time, we do replicate standard findings from the delay discounting literature such as a magnitude-of-rewards effect on discount rates. In sum, our findings suggest that circulating testosterone does not have a significant effect on delay discounting in young men.

Testosterone administration reduces lying in men.

Lying is a pervasive phenomenon with important social and economic implications. However, despite substantial interest in the prevalence and determinants of lying, little is known about its biological foundations. Here we study a potential hormonal influence, focusing on the steroid hormone testosterone, which has been shown to play an important role in social behavior. In a double-blind placebo-controlled study, 91 healthy men (24.32±2.73 years) received a transdermal administration of 50 mg of testosterone (n=46) or a placebo (n=45). Subsequently, subjects participated in a simple task, in which their payoff depended on the self-reported outcome of a die-roll. Subjects could increase their payoff by lying without fear of being caught. Our results show that testosterone administration substantially decreases lying in men. Self-serving lying occurred in both groups, however, reported payoffs were significantly lower in the testosterone group (p<0.01). Our results contribute to the recent debate on the effect of testosterone on prosocial behavior and its underlying channels.