It is my impression, and that of my colleagues, that transport and land use researchers, especially those in the accessibility community, tend to cite more nearby researchers and fewer far away researchers.
That is, European accessibility researchers cite Europeans far more than North Americans, and vice versa, independent of relevance and appropriateness. Canadians cite Canadians more than Americans, and vice versa. In short, citability decays with distance. Famous papers from across the ocean might be cited, but third tier local papers will also get cited.
This is a generalization of the self-citation observation, that people tend to cite their own previous work more than that of others, all else equal.
It can be argued this is based on proximity, certainly self-citation is (what is more proximate than my own brain), and citing locals can be presumably justified based on attending conferences and lectures, which is more likely if you are local. I think it is more based on familiarity and friendship and alliance, and if I help their career they will help mine.
Citations should be largely independent of origin. In a modern world with all of the world’s research at your fingertips, there is little excuse for being unfamiliar with far away research published in respectable journals.
I have seen research on co-citation, but nothing on geographic proximity (which doesn’t mean it doesn’t exist, but it doesn’t pop up quickly), which makes sense given that it would be a lot more work to locate the authors of cited papers.
Update, hypothesis corroborated: See World citation and collaboration networks: uncovering the role of geography in science (on arXiv, not yet in a journal)
While someone can do a rigorous geo-spatial analysis, this is a blog post, not an academic paper, so I will pose this as a hypothesis and let someone else collect the data.