Because latent learning is done when there isn't an actual reward presented...there is no reinforcement. So, for example you put a rat in a maze and let it run through the maze, but present no reward once the rat completed the maze...Then, say after like 30 times of the rat running the maze without a reward, you present cheese on the 30th trial at the end of the maze. This rat who learned the maze without a reward previously will reach the cheese (reward) extremely quickly since it already learned the maze!
Positive reinforcement, which can be positive or negative, can be demonstrated as so - a rat completes a maze during a trial, then reward with cheese that trial. Repeat this every trail. This would be positive reinforcement because you are adding something (cheese) to allow the rat to continue with the behavior (completing the maze).
Sorry if this wasn't explained super well, but I'm sure khan has some videos on this too.