| Summary: | Governments use taxes
to discourage undesired behaviors and encourage desired ones. One target of
such interventions is reckless behavior, such as texting while driving, which
in most cases is harmless but sometimes leads to catastrophic outcomes. Past
research has demonstrated how interventions can backfire when the tax on one
reckless behavior is set too high whereas other less attractive reckless
actions remain untaxed. In the context of experience-based decisions, this
undesirable outcome arises from people behaving as if they underweighted rare
events, which according to a popular theoretical account can result from basing
decisions on a small, random sample of past experiences. Here, we reevaluate
the adverse effect of overtaxation using an alternative account focused on
recency. We show that a reinforcement-learning model that weights recently
observed outcomes more strongly than than those observed in the past can
provide an equally good account of people's behavior. Furthermore, we show that
there exist two groups of individuals who show qualitatively distinct patterns
of behavior in response to the experience of catastrophic outcomes. We conclude
that targeted interventions tailored for a small group of myopic individuals
who disregard catastrophic outcomes soon after they have been experienced can
be nearly as effective as an omnibus intervention based on taxation that
affects everyone.
|
|---|