ÌÇÐÄTV

The vast amounts of public data available from social media networks, government agencies, and other internet sources present exciting new possibilities for posing questions about how people make decisions in the real world. In this talk I discuss two different lines of work which highlight how the analysis of large naturalistic datasets—or so-called ‘big data’—can be used to enrich theories of decision-making in psychology, neuroscience, and behavioral economics which are traditionally rooted in, but often constrained by artificial laboratory experiments. For example, a recent body of work suggests that fluctuations in mood states are driven by unpredictable outcomes in daily life, but also appear to drive consequential behaviors such as risk-taking. We examine how real-world unexpected outcomes predict mood state changes observable at the level of a city. By analyzing analyzing day-to-day mood language extracted from 5.2 million location-specific ‘tweets’, we find that real-world ‘prediction errors’—local outcomes that deviate positively from expectations—predict day-to-day mood states observable at the level of a city, which in turn predicted increased per-person lottery gambling rates. These results elucidate the interplay between prediction errors, moods, and risky decision-making in the real world. In a separate line of work, we examine real-world evidence for the context-dependent nature of choice—which means that the value of an option depends not only on the option in question, but also on the other options in the choice set (the ‘context’)— by analyzing a massive restaurant rating dataset (Yelp.com; 4.2 million ratings). We find that users make fewer ratings-maximizing choices in choice sets with higher-rated options—a hallmark of context-dependent choice— and that post-choice restaurant ratings also varied systematically with the ratings of unchosen restaurants. These context effects are well characterized, computationally, by a divisive normalization model of choice, in which subjective expectations of options’ quality are computed in a context-dependent manner. Taken together, these lines of work underscore how real-world, naturalistic datasets afford unique understanding of consequential choice behaviors.