Learning in honeybees as a function of amount of reward: Control of delay