Lie detection in stated preferences: The reward and recoding approaches