Traditional end-of-chapter problems often are localized, requiring formulas only within a single chapter. Students frequently can solve these problems by performing "plug-and-chug" without recognizing underlying concepts. We designed open-ended problems that require a synthesis of concepts that are broadly separated in the teaching time line, militating against students' blindly invoking locally introduced formulas. Each problem was encapsulated into a sequence with two preceding conceptually-based multiple-choice questions. These conceptual questions address the same underlying concepts as the subsequent problem, providing students with guided conceptual scaffolding. When solving the problem, students were explicitly advised to search for underlying connections based on the conceptual questions. Both small-scale interviews and a large-scale written test were conducted to investigate the effects of guided conceptual scaffolding on student problem solving. Specifically, student performance on the open-ended problems was compared between those who received scaffolding and those who did not. A further analysis of whether the conceptual scaffolding was equivalent to mere cueing also was conducted.