Dynamic Decision Making in Complex Task Environments: Principles and Neural Mechanisms
Abstract
We investigated the nature of human decision making in the leaky competing accumulator (LCA) model, a model that links behavior to underlying neural activity through abstract dynamical models in which noisy evidence is accumulated toward a decision, subject to leak and inhibition. The model predicts that evidence coming at different times can receive differential weight, and decisions outcomes should exhibit a mixture of discreteness and continuity. These predictions were confirmed across several studies using more complex task settings than previous studies. Additional studies investigated the optimization of decision making within a simplified reduction of the LCA, while others uncovered evidence establishing how differential payoff information affects decision making. Still other studies investigated the brain basis of decision making, using neuronal recording methods and noninvasive human brain activity measurement methods. Findings from these studies revealed that reward information can change the starting point of evidence integration and that evidence selection and integration are interwoven in the process of reaching a decision.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 2013
- Accession Number
- ADA581230
Entities
People
- James McClelland
- Jochen Ditterich
- Philip Holmes
- William Newsome
Organizations
- Stanford University