Sure.

Any mining method represents a competition, the aim of which is to decide which network node has the right (or priority) to mint the next block.

When a Hive block is minted, a proof-of-play gameplay opportunity is created; the block hash of the Hive block is used as a game seed. In the case of the initial tile-placing game, the seed serves to shuffle the initial tile stack (which then gets subsequently shuffled every time the player places a tile, depending on the placed tile's position and rotation).

The output of all this is a vector which tracks the decisions made by the player, and allows the validation code to fully replay their game. So proof-of-play is a function of this seed hash, and the user's gameplay. Giving these as inputs to a heuristic scoring function, f(seed hash, gameplay) = score, we have a way of checking whether the player performed well enough to win the right to mint the block.

So, when a player completes the game, if the score meets a required target, and if nobody else has yet claimed the game opportunity created by the same specific Hive block, their wallet checks the mempool, assembles any waiting transactions into a block, and submits the block.