If you watch birds for long enough you see them doing silly things that seemingly have no evolutionary benefit. For example, I once saw a magpie, upside down on the floor with its legs in the air. My first thought is that it was hurt, but then soon enough another bird came and started play fighting with it and it just was taunting with its legs at the other bird, and then eventually flipping right side up and zooming off with the other in pursuit. Perhaps ground fighting is a useful skill, or perhaps birds also experience 'fun' from novel experiences. To flip it around, it's been posited ever since Darwin that a human predilections to take on a life drawing or woodworking course is driven by sexual selection.

But without any external means to distinguish a conscious and unconscious agent, whether animals experience qualia is in a philosophical sense as unknowable as whether other humans experience qualia. Everyone except one's self could be 'philosophical zombies'[0]. This lack of verifiability is part of what is termed the 'hard problem of consciousness'[1]. If we ever build an AI agent that is conscious, we wouldn't know.

Common sense would dictate, that because I, as a human, am conscious, any other human is most certainly conscious too. But if play behaviour isn't enough to hold this certainty, where should the bar be set?

[0] https://en.wikipedia.org/wiki/Philosophical_zombie

[1] https://en.wikipedia.org/wiki/Hard_problem_of_consciousness