Your question is Exploration vs Exploitation. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
How do reinforcement learning agents balance exploration and exploitation? Describe a scenario where you would choose a model-based RL approach over a model-free approach.