Overharvesting in human patch foraging reflects rational structure learning and adaptive planning.

Area of Science:

• Behavioral Ecology
• Cognitive Science
• Computational Neuroscience

Background:

• Patch foraging involves deciding whether to stay with a resource or seek a new one.
• Organisms often deviate from optimal foraging strategies, exhibiting
• overharvesting
• behavior by staying too long.

Purpose of the Study:

• To investigate the underlying mechanisms of systematic deviations from optimal foraging strategies.
• To model foraging behavior as a problem integrating decision-making and learning under uncertainty.

Main Methods:

• Developed a computational model where foragers infer environmental structure and use uncertainty to discount future rewards.
• Conducted experiments with human participants in a patch-leaving task to observe foraging decisions.
• Compared empirical data with model predictions to validate the proposed mechanism.

Main Results:

• The model demonstrates that rational statistical inference and uncertainty adaptation can lead to overharvesting.
• Human participants' foraging behavior adapted to environmental richness and dynamics, aligning with model predictions.
• Overharvesting is explained as a consequence of adaptive uncertainty management in foraging.

Conclusions:

• Optimal foraging theory may need expansion to incorporate how foragers manage uncertainty about their environment.
• The proposed model offers a unified framework for understanding foraging decisions as adaptive learning processes.
• Findings highlight the role of statistical inference and uncertainty in shaping foraging behavior across species.