Competitive integration of time and reward explains value-sensitive foraging decisions and frontal cortex ramping...

Area of Science:

• Neuroscience
• Animal Behavior
• Computational Neuroscience

Background:

• Patch foraging is a fundamental behavior where animals must decide when to leave a depleting resource patch.
• Optimal foraging theory provides a framework for understanding these decisions, but real-world behavior often deviates.
• Understanding the neural mechanisms underlying these decisions is crucial for explaining animal behavior.

Purpose of the Study:

• To investigate the behavioral mechanisms and neural correlates of patch foraging decisions in mice.
• To develop computational models that capture realistic foraging behavior, including deviations from optimality.
• To identify the brain regions and neural dynamics involved in integrating temporal and reward information during foraging.

Main Methods:

• Developed a virtual foraging task to systematically manipulate patch value and observe mouse behavior.
• Constructed computational models integrating time and reward information, modulated by a latent patience state.
• Utilized Neuropixels recordings to capture neural activity across frontal brain areas during the foraging task.

Main Results:

• Mouse foraging behavior systematically varied with patch value and could be explained by models integrating time and rewards.
• These models quantitatively captured deviations from optimal foraging predictions.
• Distributed ramping neural signals were observed in the frontal cortex, correlating with model decision variables.
• These neural signals exhibited characteristics consistent with integration of time, reward, and memory.

Conclusions:

• Frontal cortex ramping dynamics represent a plausible neural mechanism for solving patch foraging problems.
• The findings provide a mechanistic explanation for foraging decisions, moving beyond purely normative models.
• This study links computational models of decision-making to specific neural implementations in the brain.