Figural effect is a response bias that depends on the arrangement of the terms in syllogisms. The usual pattern in construction tasks is that participants tend to give A-C responses when the syllogisms are in figure 4(A-B/B-C), while the majority of their conclusions are in the opposite direction (C-A) for syllogisms in figure 1 (B-A/C-B). We report on one experiment that examines the role of the presentation format of premises (two-line presentation vs. one-line presentation) in categorical syllogisms. The experiment showed that when premises were presented in one-line presentation format, participants generated more A-C than C-A conclusions in all syllogistic figures. However, when premises were presented in twoline presentation format, participants generated more A-C conclusions in figure 4 and more C-A conclusions in figure 1, while there were no reliable differences for either figure 2 (A-B/C-B) or figure 3(B-A/B-C). We discuss the implications of these results for different theories of reasoning.