Deconstructing risk: separable encoding of variance and skewness in the brain

Area of Science:

• Neuroscience
• Decision Science
• Behavioral Economics

Background:

• Risk assessment in decision-making is complex, involving evaluation of potential outcomes and individual risk tolerance.
• Traditional models often quantify risk using statistical variance alone, potentially overlooking other critical factors.

Purpose of the Study:

• To investigate whether human choice behavior is sensitive to both the dispersion (variance) and asymmetry (skewness) of risky outcomes.
• To identify the neural correlates associated with processing variance and skewness during risky decision-making using functional magnetic resonance imaging (fMRI).
• To explore how objective risk metrics are integrated with subjective risk preferences in the brain.

Main Methods:

• Development of a novel behavioral paradigm to independently manipulate outcome variance and skewness.
• Utilizing fMRI to scan human subjects while they made choices involving gambles with varying statistical properties.
• Analyzing neural activity in relation to behavioral choices and the manipulated risk parameters.

Main Results:

• Behavioral analysis confirmed sensitivity to both variance and skewness in risky choices.
• fMRI data revealed dissociable neural representations: parietal cortex for variance and prefrontal cortex/ventral striatum for skewness.
• Anterior insula activity showed subject-specific coupling with behavior, reflecting the integration of risk metrics and individual preferences.

Conclusions:

• Risk perception and decision-making are not monolithic, being influenced by distinct statistical properties of outcomes.
• Neural processing of risk involves segregated brain regions for different risk components (variance vs. skewness).
• The anterior insula plays a crucial role in integrating objective risk information with subjective preferences to guide behavior.