This article is based on conclusions from a survey conducted through a company called Lucid, which does not release any of its data online. Some of the authors may have released more information, I haven't been able to find anything, perhaps it's merely pending publication.
So my answer will simply presume that the information is correct and well-controlled for biases, and it was written by academics whose focus is attempting to answer these kinds of questions, so it's a reasonable thing to presume.
The question of whether or not people vote based on party affiliation or knowledge and consideration of policy is not limited to the US. In fact one of the authors of this article, Eric Guntermann, has a paper under review titled Policy Preferences Do Influence Vote Choice: Evidence from the 2017 French Presidential Election in which they write
Scholars have long debated whether policy preferences motivate vote
choice, as expected in models of policy representation. This debate
has persisted for so long because nearly all analyses of vote choice
are liable to the critique that the policy preferences in question are
endogenous to party preferences. [...] We conclude that policy
preferences clearly do matter to vote choice and that this effect is
most visible when a new party emerges.
So in the absence of information, people presume that the views of their political party aligns with their own views. The authors imply that as election day approaches, people will become more and more informed (citation needed, though it seems a reasonable assumption), and that this will likely have an impact on voter choice:
Since voters increasingly pay attention to the presidential race as
Election Day approaches, Mr. Trump has reason to worry that voters
will learn more about his stances.
Casual reading could infer from this article that US citizens are completely irrational, but I would argue that the idea is instead that when you are faithful to a party, and you do not know its position on a question, you will not pay much attention and assume by default that the party agrees with you. Until the day you realize that it doesn't. This day is much more likely to happen the closer you get to an election. There is nothing uniquely american about this process.
There is another component that argues that affiliation with a party influences your policy beliefs, and that you will change your beliefs to conform with the party you adhere to, particularly in certain contexts. You can find more information about that in a previous paper by Eric Guntermann.
Study 2, a survey experiment in Galicia, shows that party cue effects
only occur when participants are exposed to competing cues from their
preferred party and from a disliked party. Parties thus influence
opinions when they adopt contrasting positions even on issues that are
rooted in identity.
For evidence that the specific cognitive bias of projection, or assuming that the party you support agrees with your values, occurs in other parts of the world, I recommend reading more about assimilation and contrast effects in voter projections. Assimilation is when you presume that a political party you like will agree with you, and constrast is when you presume that a political party you don't like will disagree with you. The study I linked is a bit old (2001) but shows this effect in three countries, and now that you know what it's called (more than just projection), you can easily find more example for other parts of the world.