Extending a biologically inspired model of choice: multi-alternatives, nonlinearity and value-based multidimensional...

Area of Science:

• Computational Neuroscience
• Cognitive Science
• Decision Neuroscience

Background:

• The leaky competing accumulator (LCA) is a prominent model for understanding choice behavior.
• It explains neuronal accumulation and competition dynamics relevant to decision-making.
• The LCA model successfully accounts for reaction time distributions in psychophysical experiments.

Purpose of the Study:

• To analyze and extend the leaky competing accumulator (LCA) model.
• To investigate conditions for optimal performance within the LCA framework.
• To explore the impact of nonlinearities on choice behavior and value-based decisions.

Main Methods:

• Review of LCA model dynamics and performance optimization conditions.
• Simulation of LCA with neural-type nonlinearities for multi-alternative choice.
• Extension of the LCA to value-based choice scenarios, incorporating nonlinear value functions.

Main Results:

• Identified conditions under which the LCA model achieves optimal performance.
• Demonstrated that neural nonlinearities enhance decision accuracy with increased choice alternatives.
• Showcased how nonlinear value functions explain risk aversion and preference reversals.

Conclusions:

• The LCA model provides a robust framework for choice behavior.
• Nonlinearities are crucial for explaining complex decision-making phenomena, including risk and multi-attribute choices.
• The extended LCA model offers insights into value-based decision processes and their neural underpinnings.