This paper addresses the key assumption in behavioral and transportation planning literature that, when people use a transit system more frequently, they become less dependent on and less sensitive to transit maps in their decision-making. Therefore, according to this assumption, map changes are much less impactful to travel decisions of frequent riders than to that of first-time or new passengers. This assumption-though never empirically validated-has been the major hurdle for transit maps to becoming a planning tool to change passengers\' behavior. This paper examines this assumption using the Washington DC metro map as a case study by conducting a route choice experiment between 30 Origin-Destination (O-D) pairs on seven metro map designs. The experiment targets two types of passengers: frequent metro riders through advertisements on a free daily newspaper available at DC metro stations, and general residents in the Washington metropolitan area through Amazon Mechanical Turk, an online crowdsourcing platform. A total of 255 and 371 participants made 2024 and 2960 route choices in the respective experiments. The results show that frequent passengers are in fact more sensitive to subtle changes in map design than general residents who are less likely to be familiar with the metro map and therefore unaffected by map changes presented in the alternative designs. The work disproves the aforementioned assumption and further validates metro maps as an effective planning tool in transit systems.