Disentangling computational and neural mechanisms of evaluation direction in value-based decision-making

Area of Science:

• Cognitive Neuroscience
• Decision Science
• Neuroeconomics

Background:

• Value-based decision-making research often assumes preference for high-value options.
• Information evaluation can dynamically shift towards low-value evidence during decision-making.
• Prior studies have not fully distinguished these evaluation modes and their underlying dynamics.

Purpose of the Study:

• To investigate the cognitive and neural signatures of shifting evaluation directions in value-based decision-making.
• To disentangle the distinct processes associated with high-value versus low-value directed evaluations.
• To explore the temporal dynamics of value encoding during decision episodes.

Main Methods:

• Recruited 36 participants for a value-based binary choice task.
• Manipulated evaluation direction using task framing.
• Utilized behavioral measures, a hierarchical drift diffusion model (HDDM), and electroencephalography (EEG).

Main Results:

• Low-value evaluations showed longer reaction times, slower evidence accumulation, and higher decision thresholds (increased deliberation).
• EEG revealed enhanced N200 and centro-parietal positivity (CPP) amplitudes in the low-value condition, indicating value conflict and reduced integration efficiency.
• Increased alpha and beta desynchronization suggested greater attentional demand and decision commitment during low-value evaluations.

Conclusions:

• Value evaluation direction significantly modulates decision-making processes, affecting evidence accumulation and action preparation.
• Distinct cognitive and neural mechanisms underlie shifts between high-value and low-value evidence evaluation.
• The findings provide insights into flexible value encoding and a new framework for analyzing value construction in decision-making.