Value-complexity tradeoff explains mouse navigational learning.

Area of Science:

• Behavioral Neuroscience
• Computational Neuroscience
• Animal Behavior

Background:

• Understanding animal navigation and learning is crucial in neuroscience.
• Existing models often focus on single metrics, not the interplay of factors.
• The Morris Water Maze is a standard task for studying spatial learning.

Purpose of the Study:

• To introduce a novel framework for quantifying animal behavior using value and complexity.
• To model the Morris Water Maze task to analyze learning dynamics.
• To investigate how this framework can differentiate between animal groups.

Main Methods:

• Developed a dynamical system model for the Morris Water Maze.
• Solved optimal control problems under varying complexity constraints.
• Defined trajectory value (energetic cost, swimming time) and complexity (deviation from naive behavior).

Main Results:

• The model analytically describes the learning process.
• Learning is characterized by initial value optimization followed by complexity reduction.
• Complexity metric effectively distinguishes between different mouse strains' behaviors.

Conclusions:

• A novel value-complexity tradeoff framework accurately describes animal navigation learning.
• This approach provides a sensitive measure for behavioral phenotyping.
• The findings offer new insights into the computational principles of motor learning.