From high- to one-dimensional dynamics of decision making: testing simplifications in attractor models

Area of Science:

• Cognitive Science
• Computational Neuroscience
• Decision Science

Background:

• Computational models of decision making often simplify complex systems by representing subprocesses within a single dynamic description.
• This simplification assumes overall dynamics are independent of specific sub-process values.

Purpose of the Study:

• To empirically validate the simplification in attractor models of decision making.
• To investigate choice perseveration in a nonverbal, value-based decision task.
• To determine if choice perseveration is independent of attribute dimension.

Main Methods:

• Empirical investigation of choice perseveration in a nonverbal, value-based decision task.
• Testing the hypothesis that choice perseveration occurs irrespective of the attribute dimension.
• Utilizing a computational model simulation to guide empirical testing.

Main Results:

• Evidence was found supporting the validity of the simplification in attractor models.
• Choice perseveration was observed to be independent of the attribute dimension.
• The study's findings align with the computational model's simulation predictions.

Conclusions:

• The simplification in attractor models, representing overall dynamics as a single variable, is empirically supported.
• This simplification may capture essential mechanistic aspects of decision-making processes.
• The study validates a common approach in computational modeling of decision making.