The Description–Experience Gap in Risky and Ambiguous Gambles

Abstract

Recent research in decision making reported a description-experience (DE) gap: opposite risky choices when decisions are made from descriptions (gambles in which probability distributions and outcomes are explicitly stated) and when decisions are made from experience (the knowledge of the gambles is obtained by sampling outcomes from unknown probability distributions before making a choice). The DE gap has been reported in gambles commonly involving a risky option (outcomes drawn from a fixed probability distribution) and a safe option (probability of the outcome is 1), or in gambles involving two risky options. Here, we extend the study of the DE gap to gambles in which people choose between a risky option and an ambiguous option (with two nested probability distributions, where the eventgeneration mechanism is more opaque than that in the risky option). We report empirical evidence and show a DE gap in gambles involving risky and ambiguous options. Participants' choices are influenced by the information format and by the ambiguous option: participants are ambiguity-seeking in experience and ambiguity-averse in description in problems involving both gains and losses. In order to find reasons for our results, we investigate participants' sampling behavior, and this analysis indicates choices according to a cognitive model of experiential decisions (instance-based learning). In experience, participants have small sample sizes, and participants choose options where high outcomes are experienced more frequently than expected. We discuss the implications of our results for the psychology of decision making in complex environments.

Overcoming Ambiguity Aversion Through Experience

“…Our findings regarding the level of ambiguity aversion are broadly consistent with the conclusion of Dutt et al ()—experience reduces ambiguity aversion—but our reasons for reaching this conclusion are different. First of all, we characterized ambiguity with multiple and unknown probability distributions (i.e.…”

“…However, such a difference in the magnitude of the levels of ambiguity aversion is not unexpected because of the difference in sampling methods (outcome sampling versus distribution sampling) and hence the information experienced. Experiencing uncertainty about both probabilities and outcomes (as in our study) rather than only uncertain outcomes [as in Dutt et al (2013) study] might lead some people to be more cautious about the ambiguous box in our study, resulting in an ambiguity neutral pattern (i.e. around half of participants selecting the ambiguous box) instead of ambiguity-seeking.…”

“…The formulation of the ambiguous box in the Ambiguous group of the Experience condition was completely different from that of the other SOP groups as well as from those in the previous literature [i.e. Dutt et al (2013)]: There were three potential probability distributions (a uniform, a normal, or a binomial) that could underlie the ambiguous box, and in each sampling trial, one of them was randomly assigned. This formulation was in line with the theoretical definition of ambiguity, that is the inability to assign a single probability distribution to a specific outcome probability (see Camerer & Weber, 1992, p. 332) since there were multiple potential underlying distributions and it was unknown to the participants which distribution will come up in which sampling trial (and in the final choice stage).…”

“…corresponding to our "Equal Probability" case presented in Figure 1). Their results demonstrate that "experiencing" the win/lose outcomes from both risky and the so-called ambiguous boxes through the outcome sampling method decreased ambiguity aversion and resulted in an ambiguity-seeking pattern [Dutt et al, 2013; also see Ert & Trautmann (2014) for another study investigating a similar issue in an Ellsberg type task through an outcome sampling method and showing that ambiguity aversion reduces when FOP was 0.5]. However, in accordance with our aims, we took one further step in incorporating ambiguity into the sampling experience paradigm, and thus, instead of an outcome sampling method, we used a distribution sampling method.…”

Framing and Bias: A Literature Review of Recent Findings

“…According to numerous authors (e.g. Dutt et al, 2013;Gonzalez, 2013;Harman and Gonzalez, 2015), most of the biases created during the first contact with a description of a situation are weakened or completely disappear if the selection is based on experience. Huerta et al (2012) describe this phenomenon as preferences developing by experience.…”