The Korean College Scholastic Ability Test (CSAT) aims to assess Korean high school students' scholastic ability required for college readiness. As a high-stakes test, the examination serves as a pivotal hurdle for university admission and exerts a strong washback effect on the educational system in Korea. The present study set out to investigate how highly proficient Korean English as a foreign language (EFL) college readers construct meaning from two variations of texts: the lengthened and original versions of fill-in-the-blank multiple-choice question (FB-MCQ) passages extracted from the English reading section of the CSAT. A cohort of 54 highly advanced Korean EFL college students was randomly divided into groups assigned to either the original CSAT-length passages or their longer counterparts. After answering an FB-MCQ for each passage, the participants were tasked to perform an immediate free recall. Results revealed that longer texts facilitated the readers' retention of textual content and organization at the macro-structural levels, whereas the original-length group encountered challenges in arranging propositions in a hierarchical and coherent sequence and generated incorrect inferences much more frequently. Implications about the validity of the CSAT's use of brief expository passages are discussed.