Janet Metcalfe - US grants

Despite frequent and vivid reminders of the destruction that natural disasters can leave in their wake, compliance rates with evacuation orders and support for mitigation efforts remain troublingly low. One possible contributing factor to this apparent under-weighting of the risk from environmental hazards is the fact that when making decisions between uncertain outcomes in laboratory studies, people appear to treat rare events differently if their probabilities are learned through sampling and feedback (experience), as opposed to summary statistics (descriptions). Despite a recent surge in laboratory research on this topic, it is not yet clear how, or whether, this effect applies in the far more complex environment of real-world decisions about hazard preparedness. The current research project systematically investigates how learning about environmental hazards through personal experience affects people's perceptions of (and therefore their actions in response to) such dangers. The results provide a framework for scientists and policymakers to better understand how the public's personal experience with those natural hazards might interact with -- or override -- the descriptive warning information they disseminate. Appreciation of the psychological hurdles to the understanding or use of such information can improve the influence of hazard forecasts, increasing compliance with warnings and protective measures.

The Description-Experience (DE) Gap in risky choice describes the shift from Prospect Theory's predicted over-weighting of rare events when probabilities are learned through summary statistics (description), to an apparent under-weighting of small probabilities when learning occurs through repeated feedback (experience). This research program develops methods to extend existing findings about the DE gap into laboratory paradigms that more closely address the complexities inherent to real-world environmental decisions such as how to prepare for potential hazards. The research then validates existing findings through field studies that examine decisions from real-world experiences and assess the validity of prior results. Field studies integrate choice problems used in the laboratory studies into realistic paradigms, and test the boundaries of the type, timing, and valence of feedback typically used in DE gap studies. By converging evidence from well-controlled laboratory tasks and the richer, more complex set of real-world choices that people face every day, this program of research addresses current theory more thoroughly and effectively than either approach could alone. Insight into how provision of probability and outcome information in different ways affects real-world decisions illuminates ways of communicating information about environmental hazards to increase the likelihood that people will take protective action.

Learners are more engaged when curious. They are less likely to mind wander; and they are more likely to seek answers, to be attentive and ready to learn, and they actually learn better as evidenced through both brain and behavioral outcomes. However, little is known about either the neural basis of curiosity, or ways to induce it. This project proposes that curiosity is a metacognitive state consisting of feelings that people almost know the answer to the object of curiosity. This project proposes a series of experiments that investigate both cognitive/behavioral and brain responses. It tests the novel hypothesis that curiosity can be induced by manipulations that alter metacognitive processes, with favorable consequences for learning. Understanding the scientific basis of curiosity will have broad practical and educational implications for increasing students' learning and their desire to learn.

Past work on curiosity is largely correlational, and often confounds a priori knowledge with feelings of curiosity. In contrast, the experiments in this project induce curiosity via manipulations that alter metacognitive states. Behavioral and electrophysiological experiments, will investigate the impact of curiosity on attention and learning. In these experiments, participants will be exposed to computer-controlled learning scenarios where curiosity is manipulated via priming, fluency, multitasking, divided attention or inhibitory retrieval cuing. Dependent variables include reaction and study times, recall scores, and event-related potentials, as well as ratings of curiosity, confidence, and mind wandering. Curiosity will be manipulated using methods known to influence metacognitive processes. The electrophysiological studies investigating neural processes will allow use of mediational models to evaluate the role of distinct behavioral and brain components related to curiosity. These experiments will disentangle learning attributable to a priori knowledge versus those attributable to curiosity states. This project will help us to understand foundational neural and behavioral processes underlying curiosity-related learning and reward in learning, and the role of metacognition. There is potential to develop practical methods for systematically inducing curiosity which can be applied to classroom and other settings to enhance learning.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.