Log in / Register
Home arrow Computer Science arrow Social Computing, Behavioral-Cultural Modeling and Prediction
< Prev   CONTENTS   Next >

Incorporating Interpretation into Risky Decision-Making A Computational Model

Abstract. Most leading computational theories of decision-making under risk do not have mechanisms to account for the incorporation of cultural factors. Therefore, they are of limited utility to scholars and practitioners who wish to model, and predict, how culture influences decision outcomes. Fuzzy Trace Theory (FTT) posits that people encode risk information at multiple levels of representation – namely, gist, which captures the culturally contingent meaning, or interpretation, of a stimulus, and verbatim, which is a detailed symbolic representation of the stimulus. Decision-makers prefer to rely on gist representations, although conflicts between gist and verbatim can attenuate this reliance. In this paper, we present a computational model of Fuzzy Trace Theory, which is able to successfully predict 14 experimental effects using a small number of assumptions. This technique may ultimately form the basis for an agent-based model, whose rule sets incorporate cultural and other psychosocial factors.

Keywords: framing, gist, verbatim, cultural modeling.

1 Introduction

Anthropologists, such as Mary Douglas (e.g., [1]) argue that a group member's perception of risk is driven by cultural norms that define that group's identity. Similarly, Jasanoff (e.g., [2]) and others in the field of Science and Technology Studies, argue that risk is a social construct that is group-based. In contrast, scholars such as Sunstein (e.g., [3]) argue that risks are objective, and must therefore be addressed in a manner consistent with known statistics. In this paper, we draw upon Fuzzy Trace Theory (FTT; e.g., [4]), a theory of decision-making under risk, which explicitly acknowledges that risk perception contains elements of subjective perception that are shaped by culture, emotion, and prior experience. Our goal here is to formalize FTT, thereby generating a computational theory that can be used to predict the outcomes of risky decisions given the gists that are held by a member of a given group. We will therefore specify the theory to such an extent that it may be used as a set of rules, e.g., for a population of agents within an agent-based model. To that end, we have generated a novel computational representation of FTT, which is described in Section 2.

Adherents of the expected utility paradigm claim that a rational decision-maker would choose between two risky options based upon which option yielded the largest expected payoff. Nevertheless, previous studies, most notably those of Tversky and Kahneman ([5]), have demonstrated the existence of consistent heuristics and biases in human decision making, using scenarios such as what has become known as the Disease Problem (DP) in the decision-making literature:

“Imagine that the U.S. is preparing for the outbreak of an unusual Asian disease, which is expected to kill 600 people. Two alternative programs to combat the disease have been proposed. Assume that the exact scientific estimates of the consequences of the program are as follows:

If Program A is adopted, 200 people will be saved

If Program B is adopted, there is a 1/3 probability that 600 people will be saved and a 2/3 probability that no people will be saved.” [5]

A second framing contains the same preamble, but presents the following options: “If Program C is adopted 400 people will die.

If Program D is adopted there is a 1/3 probability that nobody will die, and a 2/3 probability that 600 people will die.”

Although subjects are more likely to choose option A than option B, they are more likely to choose option D than option C, even though these problems are mathematically identical. Adherents of the heuristics and biases approach point to Prospect Theory (PT; [6]) which weights losses and gains relative to an individual's reference point differently. PT and its successor, Cumulative Prospect Theory [7], both retained utility theory's assumption of a continuous and monotonic utility function. Such continuous models tend to be favored because they are computationally tractable.

According to Fuzzy Trace Theory (FTT), these, and similar, decision options are stored at a categorical, qualitative level of processing – known as gist – which is encoded simultaneously with the detailed verbatim numbers. “The gist of a decision varies with age, education, culture, stereotypes worldview, and other factors that affect the meaning or interpretation of information…” [8]. According to FTT, the majority of subjects in the gain frame of the DP prefer option A (D) because they interpret its decision options as:

A) Some live (die) vs.

B) Some live (die) OR none live (die)

A central tenet of FTT is that detailed representations of these numerical options are recorded, but that the gist representation is used preferentially, especially when verbatim decision-making is insufficient to distinguish between decision options. Finally, when a decision between gist categories is made, it is made on the basis of simple binary valenced affect. (e.g., good vs. bad, approach vs. avoid, etc.).

We have implemented our theory in a computer program that makes predictions regarding outcomes in risky-decision problems. The theory and its computer implementation successfully predict the outcome of 14 different experimental effects reported in the literature. We further test our theory against the outcomes of a metaanalysis of 13 studies reported in the literature. We begin with a description of the theory.

Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
Business & Finance
Computer Science
Language & Literature
Political science