Decision-making Across Development: The Impact of Ambiguity and Social Context

Public health data show that many everyday reckless behaviors reach a developmental peak in adolescence, with adolescents engaging in more reckless behaviors than both children and adults. In contrast, most studies of decision-making across development do not find laboratory risk-taking to peak in adolescence. Here, I focus on two factors that contribute to the discrepancy between public health and laboratory findings: ambiguity and social context. Everyday decisions tend to involve ambiguous decisions (choices with unknown probabilities), while previous laboratory studies have largely focused on risky decisions (choices with known probabilities). Consequently, little is known about the ambiguity preferences of young children. Across three behavioral studies, I show that ambiguity aversion is absent in 5-year-old children (Chapter 2) and 8- and 9-year-old children (Chapter 3) but present in 15- to 18-year-old adolescents (Chapter 4) and adults (Chapters 2 to 4). The results of Chapters 2 through 4 indicate that the willingness to take ambiguous gambles, like the willingness to take risky gambles, does not peak in adolescence. Everyday decisions also often occur in social contexts when friends are present and outcomes can be shared, whereas most laboratory studies occur in social isolation. In Chapter 5, I use functional magnetic resonance imaging to show that neural response to reward for self and for friend are similar in a sample of young adults (ages 18 to 28), and that neural response to reward linearly decreases with age when participants are watched by their friends but not when they are alone. In Chapter 6, I use behavioral modeling to show that adults value rewards similarly for themselves and for their friend. Adolescents, in contrast, value their own rewards more than those of their friend, but the presence of their friend reduces this valuation difference. The results of Chapters 5 and 6 indicate that the presence of friends prompts adolescents and young adults to engage in behavior that benefits both themselves and their friends. Collectively, the results in this dissertation demonstrate the need to consider contextual influences on decision-making in order to better capture everyday decision behavior in the laboratory.