Academic journal article Cognitive, Affective and Behavioral Neuroscience

Beyond Risk and Ambiguity: Deciding under Ignorance

Academic journal article Cognitive, Affective and Behavioral Neuroscience

Beyond Risk and Ambiguity: Deciding under Ignorance

Article excerpt

In this study, we examined the neural basis of decision making under different types of uncertainty that involve missing information: ambiguity (vague probabilities) and sample space ignorance (SSI; unknown outcomes). fMRI revealed that these two different types of uncertainty recruit distinct neural substrates: Ambiguity recruits the left insula, whereas SSI recruits the anterior cingulate cortex, bilateral inferior parietal cortex, and the lateral orbitofrontal cortex. The finding of unique activations for different types of uncertainty may not necessarily be predicted within the reductive approach of modern theories of decision making under uncertainty, because these theories purport that humans reduce more complicated uncertain environments to subjectively formed less complicated ones (i.e., SSI to ambiguity). The predictions of the reductive view held only for ambiguityaverse individuals and not for ambiguity-tolerant individuals. Consequently, theories of decision making under uncertainty should include individual tolerance for missing information and how these individual differences modulate the neural systems engaged during decision making. Supplemental materials for this article may be downloaded from http://cabn.psychonomic-journals.org/content/supplemental.

Despite an emerging consensus that uncertainty modulates valuation (a process, or a combination of processes, of determining the current worth of the available options; Rangel, Camerer, & Montague, 2008), most researchers focus only on two types of uncertainty: risk (known outcomes, known probabilities; von Neumann & Morgenstern, 1944) and ambiguity (known outcomes, vague probabilities; Ellsberg, 1961). The subjective expected utility (SEU) theory, prospect theory (Tversky & Kahneman, 1992), rank-dependent expected utility theory (Quiggin, 1993), and other theories treat unavailability of information as a negative characteristic of the environment and suggest that individuals reduce higher levels of uncertainty to risk by forming subjective beliefs about unknown characteristics of the environment (the reductive viewpoint)-for instance, reducing ambiguity to risk by forming subjective beliefs about the unknown probabilities. Behavioral studies have shown that individuals demonstrate ambiguity aversion (for a review, see Camerer & Weber, 1992). Platt and Huettel (2008) suggested that parameters measuring the amount of ambiguity might be encoded in the brain (in the amygdala, the orbitofrontal cortex [OFC], or the anterior insula). Other types of uncertainty, including conflict (known outcomes, conflicting information about probabilities; Smithson, 1999), unawareness (decision makers are unaware of unknown outcomes; Modica & Rustichini, 1994), and sample space ignorance (SSI; decision makers are aware of unknown outcomes; Smithson, Bartos, & Takemura, 2000), still remain understudied. Unawareness is not likely to affect valuation (Modica & Rustichini, 1999). However, conflict and SSI are as likely to modulate valuation as risk and ambiguity are, because under all these conditions decision makers are aware of the informational limitations during the valuation stage. To better understand how uncertainty modulates valuation, it is important to investigate the full variety of its types. It is also intriguing to examine whether individual preferences toward different types of uncertainty interact with their modulating effects. We begin this line of research by focusing the present article on uncertainties generated by the lack of information (ambiguity and SSI).

Baron and Frisch (1994) suggested that people regard any alternatives with missing information (MI) as inferior. Smithson et al. (2000) found that decision makers demonstrate aversion toward SSI. The reductive viewpoint suggests that individuals reduce SSI to ambiguity and then to risk by forming subjective beliefs, first about the partition of a sample space and then about the corresponding probabilities. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.