Multiple Choice Neurodynamical Model of the Uncertain Option Task.

Area of Science:

• Computational Neuroscience
• Decision Neuroscience
• Primate Neurophysiology

Background:

• The uncertain option task is used to study neural mechanisms of decision confidence.
• Previous research recorded single neuron activity in the lateral intraparietal cortex of monkeys during this task.
• The task involves choosing between a sure reward and a risky perceptual decision.

Purpose of the Study:

• To propose and validate a computational model for the uncertain option task.
• To investigate the neural basis of decision confidence and sensory-motor associations.
• To explain both behavioral and neurophysiological data from monkey experiments.

Main Methods:

• Development of a multiple choice model implemented in a discrete attractor network.
• Simulation of the model to reproduce experimental findings.
• Comparison of model predictions with existing data and alternative models.

Main Results:

• The discrete attractor network model successfully reproduced behavioral and neurophysiological data.
• The model's multistable attractor landscape explains observed monkey behavior and neural activity.
• The model generates testable predictions, including one to differentiate it from continuous attractor models.

Conclusions:

• The proposed discrete attractor model provides a robust explanation for the uncertain option task.
• Findings support the interpretation of the uncertain option task as a sensory-motor association.
• The model offers a framework for understanding decision confidence and generates novel experimental predictions.