History education scholars have recognized the need for test validity research in recent years and have called for empirical studies that explore how to best measure historical thinking processes. The present study was designed to help answer this call and to provide a model that others can adapt to carry this line of research forward. It employed think-aloud protocols with 12 high school history students to explore whether selected multiple-choice items from a new test of historical thinking--the "Historical Thinking Test" (HTT)--elicit the intended aspects of historical thinking among high school history students and whether these new items elicit these constructs at higher rates than items from extant standardized history tests. Data revealed that HTT multiple-choice items did, to varying degrees, elicit aspects of historical thinking and that HTT items elicited the intended aspects of historical thinking at higher rates than selected multiple-choice items from extant standardized tests. Yet the HTT items also elicited construct-irrelevant reasoning, which weakens an argument for their use as measures of historical thinking. Implications for history testing and the efficacy of multiple-choice items as measures of historical thinking are discussed.