Informational Entropy Threshold as a Physical Mechanism for Explaining Tree-like Decision Making in Humans.

Area of Science:

• Cognitive Neuroscience
• Decision Science
• Computational Psychology

Background:

• Traditional drift-diffusion models (DDM) excel at simple perceptual decisions but struggle with complex, sequential choices.
• Existing models lack mechanisms to explain decision-making involving future option prospection and uncertainty quantification.

Purpose of the Study:

• To propose and validate a novel entropy-based mechanism for understanding sequential decision-making.
• To investigate the role of internal uncertainty computation (Shannon entropy) in triggering decisions.
• To explain decision time distributions and prospection in complex tasks.

Main Methods:

• Developed a computational model using Shannon entropy to quantify uncertainty in sequential choices.
• Conducted experiments involving human participants navigating a virtual maze.
• Utilized eye-tracking technology to monitor participant behavior and prospection strategies.
• Analyzed decision times and correlated them with the entropy-based model predictions.

Main Results:

• Demonstrated that an entropy-based mechanism effectively models human decision-making in sequential tasks.
• Provided evidence that participants engage in prospection, with its level quantifiable.
• Showed that the model explains power-law tails in decision time distributions, a limitation of DDM-like frameworks.

Conclusions:

• The proposed entropy-based model offers a more comprehensive account of complex decision-making than traditional approaches.
• Internal uncertainty computation via Shannon entropy is a plausible driver for decisions in multi-option scenarios.
• The findings advance our understanding of cognitive processes in navigation and sequential choice tasks.