Zhan Guo at NYU has a very nice paper in TR part A about the distortionary effects of Harry Beck's London Underground Map: Mind the map! The impact of transit maps on path choice in public transit:
The conclusions (the paper is behind a paywall) Emphasis Added:
"This paper investigates the effect of schematic transit maps on travel decisions in public transit systems. The relationship might have significant implications for public transit operation and planning, but so far it has been largely overlooked by both academics and practitioners. The paper first defines four types of information delivered from a transit map: distortion, restoration, codification, and cognition, and then discusses their potential influence on travel location, mode, and path choices.
The case study on the London Underground confirms that a schematic transit map indeed affects passengers’ path choices. Moreover, the map effect is almost two times more influential than the actual travel time. In other words, underground passengers trust the tube map (two times) more than their own travel experience with the system. The map effect decreases when passengers become more familiar with the system but is still greater than the effect of the actual experience, even for passengers who use the underground 5 days or more per week.
The paper also shows that the codification of transfer connections is also important. Different codification can make a transfer look more or less convenient on a transit map than in reality, which will either decrease or increase the perceived transfer inconvenience for the corresponding stations. This paper observes both situations in the underground case study and quantifies this codification effect, in terms of the number of attracted or precluded transfers, for four major transfer stations: Baker St., Bank/Monument, Victoria, and Oxford Circus.
Of course, these results are only based on the London Underground, a unique case in many aspects. Few transit maps enjoy such public popularity as the tube map in London. Many transit maps include prominent geographical features, which dilute the map effect. Other systems have different past or present versions of their transit map, which precludes a lasting and stable map effect. Many metropolitan regions possess an easier-to-comprehend urban form than London, which could weaken the role of a transit map in the formation of a cognitive map. The subway map effect in New York City is probably different from that in London. Therefore, readers should be cautious about making generalizations.
If a transit map has an impact on travel decisions, what are the implications for transit operation and planning? First, if passengers trust a schematic map more than their own experience, all planning efforts aimed at changing travel behavior need to consider the map effect; otherwise, the effectiveness of those efforts might be weakened. For example, this map effect might partially explain why Advanced Traveler Information Systems (ATIS) often yields modest improvements in terms of travel time savings in public transit ([Hickman and Wilson, 1995], [Avineri and Prashker, 2006] and [Ben-Elia et al., 2008]). Secondly, a transit map might cause certain operational problems. For example, it might unintentionally shift more passengers to a congested segment in the network and thus form a bottleneck. The overcrowding at the Victoria and Oxford stations and on the link between the King’s Cross and Old Street stations, which is much shorter on the tube map than in reality, are possible examples.
Accordingly, a transit map could potentially become a planning tool to solve operational problems and improve system efficiency. For example, link lengths could be revised, and transfer stations could be re-coded on a transit map in order to change passenger behavior and mitigate platform and train crowding. Annotations of waiting time or crowding for selected stations on the map might also be important (Hochmair, 2009). Clearly, this approach has its own limits: we could not redraw a transit map however we pleased.
In terms of future trends, ATIS and alternative travel information channels, such as smart phones and the internet, might change the role of a transit map in mixed ways. On the one hand, they may weaken the transit map effect. For example, internet-based trip planners may recommend specific travel paths based on their actual attributes. On the other hand, they may strength the map influence as well. For example, a transit map might become more accessible to passengers through, for instance, smart phones or the internet. Travel information, such as crowding and delays, delivered in a map format could be more effective than other media ([Hato et al., 1999] and [Talaat, 2011]). Conventional media like the transit map will still likely be critical and indispensible for trip planning despite the prevalence of real time information (Cluett et al., 2003).
In summary, transit maps can have a profound impact on passengers’ travel decisions and system performance. Both individual passengers and transit agencies should ‘mind the map’ in order to make their best planning decisions.