One thing to consider here is that the article pulled this question out, but did not actually state how many people actually got this question right or wrong.

One thing to consider here is that the article pulled this question out, but did not actually state how many people actually got this question right or wrong. It is pretty common for these kinds of tests to include a bunch of questions along various scales, including ones that researchers expect "everyone" to get. This author might have a story they are trying to tell, and if they had information about this question like "oh well everyone got this question right", then they would not be able to illustrate what they mention afterwards: > So can unfamiliar words or concepts. It’s been estimated that comprehension starts to break down when as little as 2 percent of the words in a passage are unfamiliar. The bread-and-crackers text uses a number of words and phrases that aren’t likely to crop up in ordinary conversation, like “exposed to the elements,” “crystallized,” and “retrogradation” This seems fairly straightforward, right? Like some people will not be able to fill in the blank of the meaning of certain words, especially people who are not as good at "test taking" skills (solving the puzzle of understanding all the answers are in the text, process of elimination and other tricks that people who ace this stuff are using without even thinking).