Empirical evidence has consistently shown that interactive questions are an effective strategy for strengthening student learning through instructional videos. However, research has not directly addressed the crucial question of whether the beneficial effects of interactive questions in instructional videos are attributable to the presence of questions, or whether they result from a combination of question presence and computer-paced "interactivity." This question is important for understanding the mechanisms by which interactive questions work and holds practical significance for instructors' selection of suitable video formats for student learning. To fill this research gap, this study conducted a quasi-experiment with pre-test post-test design to determine the effects of question presence and interactivity on university students' cognitive load, emotional engagement, video view counts, and cognitive learning outcomes. Three discrete classes were randomly assigned to one of three experimental conditions: videos with questions (Experimental Group 1, n = 39), videos with interactive questions (Experimental Group 2, n = 35), and videos without questions (Control Group, n = 35). Age, gender, motivation, and actual and perceived prior knowledge were controlled in the experiment. The findings revealed that (a) question presence in traditional linear videos conferred beneficial effects on germane cognitive load, emotional engagement, knowledge acquisition, and knowledge application; (b) question interactivity did not offer additional advantages in reinforcing knowledge acquisition or application compared with question presence alone; and (c) question presence or interactivity exerted no significant effect on video view counts. This study has practical implications for instructors seeking to develop effective instructional videos.